JOURNAL OF PROPULSION AND POWER Vol. 22, No. 6, November–December 2006

Overview of United States Space Propulsion Technology and Associated Space Transportation Systems

Robert L. Sackheim∗ NASA Marshall Space Flight Center, Huntsville, Alabama 35812 DOI: 10.2514/1.23257 The space propulsion industry and, indeed, the overall space transportation industry in the United States have been in decline since the first human landing on the moon in 1969. The hoped-for reversal to this decline through the space shuttle and space station programs never really materialized. From an 80% market share in the late 1970s, the U.S. share of the world launch market fell to 20% by 2002. During that period, only two new booster engines have been developed and flight certified in this country. Only limited progress has been made in reducing engine costs or increasing performance although these factors are not necessarily directly related. Upper-stage and in-space propulsion in the United States have not fared much better in the world market. On the other hand, space-faring nations in Europe, the Middle East, Asia, and the former Soviet Union are believed to have developed 40–50 new, high-performance engines over the same period. This trend will have to be reversed to enable future exploration missions. The intent of this paper is to summarize past propulsion-system development, assess the current status of U.S. space propulsion, survey future options, evaluate potential impact of ultra low-cost, small launch-vehicle programs, and discuss some future propulsion needs for space exploration.

I. Introduction In general, space transportation and rocket propulsion technology fl HE U.S. rocket propulsion industry and associated space development in the United States for all aspects of space ight applications has significantly lagged behind the rest of the world T transportation business has been in a steady state of decline fl fi since the end of the Apollo era (1972), although the actual steep since the initial ight certi cation of the space shuttle Space funding, and associated manpower, roll-off began immediately after Transportation System (STS). This lack of progress in advancing fi rocket propulsion technologies over such a long period has resulted the successful rst human landing on the moon, Apollo 11, in 1969. fi ’ A turnaround in the propulsion and space transportation industry was in several de ciencies in today s U.S. national space program. Most notable of these is the reduced reliability in U.S. launch and space expected after the space shuttle, and even, ultimately, the space fl station, program was authorized to proceed, but the turnaround never vehicles, as evidenced by the increased number of ight failures materialized. The space shuttle program [or National Space during the late 1990s and into the new decade, as well as the large loss Transportation System (NSTS)], which had to develop three new of U.S. market share in both the space launch and spacecraft industries. As is well known, the U.S. launch market share fell from liquid rocket engines [space shuttle main engine (SSME), orbital  maneuvering engine (OME), and reaction control engine (RCE)] and 80% in the late 1970s to less than 20% worldwide in 2002. the world’s first large, segmented, and reusable solid rocket motor In the last three decades, only one new government-sponsored and one new largely commercial-sponsored booster engine have been did not reverse the decline from the Apollo era; it only slowed down fl fi the rate of decline until the late 1970s, as shown in Fig. 1. developed and gone through ight certi cation. These are the SSME and the Boeing RS-68, respectively. The SSME was, of course,

Robert L. Sackheim just recently retired from NASA, after seven years as the Assistant Director and Chief Engineer for Propulsion at NASA’s George C. Marshall Space Flight Center (MSFC). He currently is an independent consultant and an adjunct professor at the University of Alabama, Huntsville (UAH). He holds a B.S. degree from the Downloaded by Univ of Wisconsin-Madison on October 8, 2012 | http://arc.aiaa.org DOI: 10.2514/1.23257 University of Virginia, an M.S. degree from Columbia University, and has completed all doctoral course work in chemical engineering at the University of California in Los Angeles. He joined MSFC after 35 years in various technical and management positions with the TRW Space and Electronics Group. His awards and honors include the AIAA James Wyld Award for outstanding technical contributions to the field of rocket propulsion, as well as 12 NASA Group Achievement Awards. While at TRW he received three annual Chairmen’s Awards and a TRW patent of the year award. He is a fellow of AIAA and was elected in 2000 to the National Academy of Engineering. He also received the AIAA Sustained Service Award in 2000. The Alabama/Mississippi section of the AIAA awarded him the Martin Schilling Award for outstanding service to the section, the Herman Oberth Award “For Outstanding Individual Scientific Achievement in the Fields of Astronautics and Space Sciences,” and the Helgar Toftoy Award for outstanding technical management. He recently received an award from the Association of Aeronautics and Astronautics of France for “High Quality Contributions to the Propulsion Field.” In 2001 he was awarded the NASA Medal for outstanding technical leadership, and in 2003 he was awarded a presidential citation for meritorious federal executive service. He has served on a number of NASA boards, including the Shuttle Independent Assessment Team (SIAT), the Mars Climate Orbiter Mishap Investigation Board, and the Mars Polar Lander Mishap Board. He has authored more than 250 technical papers. He also holds nine patents for spacecraft and/or launch vehicle propulsion and/or control systems technology.

Received 21 February 2006; revision received 7 September 2006; accepted for publication 5 September 2006. Copyright © 2006 by the American Institute of Aeronautics and Astronautics, Inc. The U.S. Government has a royalty-free license to exercise all rights under the copyright claimed herein for Governmental purposes. All other rights are reserved by the copyright owner. Copies of this paper may be made for personal or internal use, on condition that the copier pay the $10.00 per-copy fee to the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923; include the code $10.00 in correspondence with the CCC. ∗Assistant Director and Chief Engineer for Propulsion (retired), Office of the Center Director. AIAA Fellow. 1310 SACKHEIM 1311

Apollo 11 missions. Many of these high-performance in-space propulsion First Lunar Era of the ICBMs plus Space technologies are literally required to enable many of the future space- Landing 100 Exploration and Science exploration missions planned for classes of human and robotic 90 spacecraft, as well as for many national defense missions. 80 Over 50,000 new high-tech jobs created Spawned the computer/information age 70 II. Importance and Key Functions of Space Propulsion 60 Pathfinder Space propulsion systems perform three basic but highly enabling 50 Initiation of SDIO X-vehicles the Shuttle functions for all U.S. payloads and assets that operate from space: 40 SLI, NGLT Program Orbital Space 1) They lift the launch vehicle and its payload from the Earth 30 Plane surface-based launch pad and place the payload into low Earth orbits 20 (LEO). 10 2) They transfer payloads from LEOs into higher orbits, such as

% Peak Funding for Space Propulsion Space Propulsion Funding for Peak% geosynchronous, or into trajectories for planetary encounters 1960 1970 1980 1990 2000 (including planetary landers, rovers, and sample return launchers if Fig. 1 Historical trends in national rocket propulsion funding as a required). percentage of the peak funding for Apollo by year. 3) Finally, at the mission operational location, they provide thrust for orbit maintenance, rendezvous and docking, position control, station-keeping, and spacecraft attitude control (i.e., proper pointing originally developed for the space shuttle in the 1970s. Some and dynamic stability in inertial space). significant upgrades have been incorporated since the original Each of these sets of space propulsion functions provides unique, certification for flight, but their modifications have been to increase mission-enabling capabilities. The only type of rockets capable of reliability and safety and to somewhat improve reusable launch- providing high thrust forces required for an Earth-to-orbit (ETO) vehicle (RLV) operability [i.e., increase mean time between launcher are those that convert thermal energy to kinetic energy. refurbishment (MTBR) from one mission to another, on the order of Only two such technical approaches are conceivable today: 10–15 RLV flights]. Very little advancement in reducing costs or 1) chemical combustion and 2) nuclear reactor thermal power. increasing performance has been achieved. In fact, if anything, Nuclear thermal reactor rockets are unacceptable because they are performance and cost have been sacrificed to increase safety and open-cycle devices that emit radioactive materials. Thus, chemical reliability. rockets will launch NASA payloads into LEO for the foreseeable The RS-68 engine was developed by Boeing/Rocketdyne as a future. low-cost expendable booster engine for the Delta IV evolutionary It is also important to recognize that all flight experience with expendable launch vehicle (EELV). However, from published data chemical in-space propulsion systems [Apollo, satellites, planetary by the contractor, it did not meet its original target goals in spacecraft, shuttle reaction control system (RCS), and space station] nonrecurring and recurring (unit) costs and in performance. Engine has been with storable propellants. Other than the two- or three-burn performance of the RS-68 is comparable to that of the 1960s period Centaur/RL-10, the United States has never flown an in-space Saturn V second- and third-stage J-2 engines, both of which were propulsion system that uses cryogenic propellants, even though simple open-cycle, gas-generator powered designs. The only claim many of the leading propulsion-system candidates for future for any advancement that can be made by the developer for the RS-68 exploration missions have cryogenic upper-stage systems designated is the much higher thrust level and the incorporation of modern in their conceptual designs. The long-term storage and transfer of design and manufacturing techniques. This would be very beneficial advanced cryogenic chemical propellants present a wide range of in production runs of 30–50 engines per year, but it appears that in the technical challenges that must be addressed in the near term. The near-term space marketplace only about 8–10 engines per year will nation’s ability to mature these technologies to a level that assures be needed to meet current demand. This will produce almost no unit flight readiness will be one of the key factors in ensuring the cost advantage over older engines that are available anywhere in the successful execution of future space-exploration missions. world today. A summary description of this engine is given in Sec. V. Once a spacecraft has reached LEO, high thrust is no longer While the United States has developed almost no new rocket required for in-space operations. Low-thrust systems operating for a technology during the last 30-plus years, the rest of the world has long time can achieve the same result as a short-duration, high-thrust been quite busy doing exactly the opposite. The space faring nations system. Frequently, the mission designer must trade off trip time and in Europe, Asia (including India), the Middle East, and the former propellant mass. However, for missions relating to transfer of crews

Downloaded by Univ of Wisconsin-Madison on October 8, 2012 | http://arc.aiaa.org DOI: 10.2514/1.23257 Soviet Union are believed to have developed between 40 and 50 new, from Earth to other destinations, long-trip times are not an option. high-performance engines of almost every type of combustion/ Once a spacecraft returns to a gravity well, like the moon or Mars, propellant system or power cycle that is conceivably enabled by high-thrust chemical propulsion is again required (or aerocapture as a today’s rocket technologies. Based on these observations, it is possible option if it is developed, proven, and applicable). probably no coincidence that the total U.S. share of the space launch For all of the applications inherent in human exploration of the market and U.S.-built launch-vehicle reliability has eroded badly in moon and Mars, space propulsion in general and chemical the last 40 years. In the commercial space marketplace alone, the propulsion in particular will be a technology area critical to success. United States captures only about $1 to 2 billion out of a potential Space propulsion for exploration is used to refer specifically to these worldwide commercial launch market of $8 to 10 billion per year as functions: 1) safe and reliable access to space; 2) in-space of today. maneuvering: orbit transfer, trajectory insertion, and attitude control A similar trend has also been observed in the upper-stage and in- (position control, rendezvous, and docking); and 3) ascent and space propulsion technology product development areas. Advance- descent propulsion for human operations at the moon and Mars. ments in both will be greatly needed to enable future exploration These functions are the key capabilities by which the nation’s missions. Most of the U.S. in-space propulsion developments in long-term vision for space exploration will be enabled and that must recent times have been privately funded with some support from the be made available to achieve the required exploration mission government. Even for these technology investments, most of the success. This paper will clearly and accurately illustrate that these government-sponsored projects were stopped for one reason or capabilities have been allowed to atrophy in the United States, another before any significant advances in technology readiness jeopardizing the potential success of the future planned human could be achieved. Only now has the government begun to realize exploration program. that the paucity of new technology developments for the high The mass involved with human space exploration means that the leverage of increased-performance in-space propulsion systems has primary focus of space propulsion must be on both rocket engines greatly hampered the growth of Earth-orbital and deep-space with thrust levels greater than 100,000 lbf (45,359 kgf) and in-space/ 1312 SACKHEIM

Fig. 2 U.S. rocket engine developments from 1955–2005.

lander propulsion systems with thrust levels of 25,000–50,000 lbf during the development of any new propulsion system (Table 2). (11,340–22,680 kgf). Figures 2 and 3 indicate that the United States These failures almost always result in the redesign or modification has not developed such rocket engines since the development of the and then repair and replacement of the failed propulsion hardware. SSME and the RL-10 for in-space/upper-stage vehicles (also refer to As more and more tests are performed, supported by accurate and Fig. 1). Table 1 summarizes the current capability available today for validated analytical models, design engineers gain a better upper-stage and in-space propulsion applications. Table 1 also understanding of the propulsion-system characteristics and are able reinforces the point that no significant high-thrust upper-stage or in- to reduce the frequency of test failures by applying lessons learned space propulsion systems have been developed in the last decade. and by building on the knowledge gained from previous successes The only new liquid rocket engines in this size range developed in the and failures. This development cycle allows the design team to United States and fielded after 1980 are a modification of a rocket optimize the system design, while simultaneously gaining invaluable engine developed by Russia (i.e., RD-180) and the RS-68 engine by hardware experience that will enable them to reduce the development Rocketdyne. time of future systems. This cycle of “test–fail–fix–test” also clearly demonstrates that the hardware is the key driver for development and III. Considerations for the Nation’s Space Program the associated costs. These costs must be minimized for future developments through the judicious, intelligent, and logical use of All of the lessons learned from the only four human space flight the advanced computational models and codes and modern high- programs (Mercury, Gemini, Apollo, and Shuttle) in U.S. history speed computer capabilities now available. The ideal model for this demonstrate the value of a strong, knowledgeable, highly balanced engineering approach is illustrated in Fig. 4.

Downloaded by Univ of Wisconsin-Madison on October 8, 2012 | http://arc.aiaa.org DOI: 10.2514/1.23257 experienced, and successful agency and a nationwide propulsion This observation is further substantiated by analysis of historical engineering team. The new emerging space-exploration activities rocket and jet (gas turbine) engine development costs. Close need to retain, use, and build on what remains of this previous strong examination of the detailed cost distribution for major engine propulsion technical knowledge base. development programs, compiled by the Marshall Space Flight Based on a detailed review of propulsion-system historical Center (MSFC) in the early 1990s, reveals a remarkable consistency. development data, it is apparent that hardware failures will occur This analysis was conducted for the F-1, J-2, SSME, RL-10A-3, Space-Shuttle OMS, and the F100 jet engines. The results clearly show that about 50% of the development cost is the hardware, whereas the remaining portion is evenly split between the labor elements of test, engineering, and management. The obvious conclusion is that major reductions in development costs and schedules should be achievable if the cost of the development hardware can be reduced (i.e., design the hardware and program for low cost instead of maximum performance). This observation is particularly true if the hardware is both low in cost and relatively mature (i.e., base the new design approach on existing design databases), which was not the case for the Saturn V, space shuttle, or the RL-10A-3 upper-stage engines, so that relatively few hardware failures occur. The reduction in failures will also result in fewer test cycles, further reducing costs and schedule for the engine full-scale development (FSD) program. For propulsion activities that LO =LH Fig. 3 Existing 2 2 engine production level designs. do not have past hardware development experience to build on, a Table 1 Summary of in-space rocket engines and thrusters currently in place

Type Manufacturer Propellants Nominal thrust range, Nominal Isp,s Nominal operating Engine weight, lb Status Date of development lbf (kgf) at vacuum life (kg) complete Spacecraft apogee or Aerojet, Northrop- Storables 50–300 (23–136) large Grumman, Delta-V engines IHI, AMPAC, (chemical engines/ EADS thrusters) R4-D class Aerojet, EADS MON/MMH 90–100 (41–45) 311–328 2–4 h 8 (4) Flown 1956 MMBPS class Northrop-Grumman MON/MMH 100 (45) 315 2–4 h 7.5 (3) Flight qualified 1970 (Formerly TRW) – – DMLAE Grumman-Grumman MON=N2H4 105 (48) 315 335 2 4 h 10 (4.5) Flown 1996 (formerly TRW) – DMLAE EADS MON=N2H4 100 (45) 310 2 4 h 10 (4.5) Flown 1972 – DMLAE AMPAC MON=N2H4 100 (45) 310 2 4 h 10 (4.5) Flown 1994 – DMLAE IHI MON=N2H4 100 (45) 310 2 4 h 10 (4.5) Flown 1979

Bipropellant RCS class Aerojet, AMPAC, MON/MMH 5–2.2 (2–1) 250–300 depending 1–2 h 2.5 (1) Flown 1982 SACKHEIM EADS on duty cycle – – – – SCAT Bimodal thruster Grumman Grumman MON=N2H4 14 4(6 2) 320 310 10 h 4 (2) Flown 1997 Monopropellant Grumman–Grumman, Catalytically hydrazine (N2H4) NGST, Aerojet, decomposed N2H4 AMPAC, EADS, IHI RCS-class thrusters Grumman–Grumman, Catalytically 0.1–25 (0.05–0.11) 150–230 depending 60,000 s steady-state 0.5–10 (0.23–4.5) Flown 1967 6 NGST, decomposed N2H4 on duty cycle 1  10 pulses Aerojet, AMPAC, EADS, IHI Delta-V class thrusters Grumman–Grumman, Catalytically 25–300 (11–136) 225–235 10 h 500 pulses/off 10–50 (4.5–23) Flown 1970 NGST, Aerojet, decomposed N2H4 pulses AMPAC, EADS, IHI Planetary lander class Aerojet Catalytically 500–700 (227–317) 240 5 min and throttle 20 (9) Flown 1975 decomposed N2H4 10:1 OME class Aerojet MON/MMH 11,000 (4990) 330 Reusable on shuttle 220 (100) Flown 1978 RL-10A-1 Pratt & Whitney O2=H2 15,000 (6804) 422 1200 s 300 (136) Flown 1960

Downloaded by Univ of Wisconsin-Madison on October 8, 2012 | http://arc.aiaa.org | DOI: 10.2514/1.23257 10.2514/1.23257 DOI: | | http://arc.aiaa.org 2012 8, on October Wisconsin-Madison of by Univ Downloaded RL-10A-3 Pratt & Whitney O2=H2 15,000 (6804) 427 1200 s 300 (136) Flown 1965 RL-10A-3-1 Pratt & Whitney O2=H2 15,000 (6804) 431 1200 s 300 (136) Flown 1967 RL-10A-3-3 Pratt & Whitney O2=H2 15,000 (6804) 442 1200 s 300 (136) Flown 1969 (continued) 1313 1314

Table 1 Summary of in-space rocket engines and thrusters currently in place (Continued)

Type Manufacturer Propellants Nominal thrust range, Nominal Isp,s Nominal operating Engine weight, lb Status Date of development lbf (kgf) at vacuum life (kg) complete

RL-10A-3-3A Pratt & Whitney O2=H2 16,500 (7484) 444 1200 s 300 (136) Flown 1972 RL-10A-4 Pratt & Whitney O2=H2 20,800 (9435) 449 1200 s 370 (168) Flown 1975 RL-10A-5 Pratt & Whitney O2=H2 14,560 (6604) 368 1200 s 316 (143) Flown 1978 RL-10A-4-1 Pratt & Whitney O2=H2 22,300 (10,115) 451 1200 s Flown 1982 RL-10B-2 Pratt & Whitney O2=H2 24,750 (11,226) 466.5 1200 s Flown 1998 Delta II second stage Aerojet MON/A-50 10,000 (4536) 315 30 min 220 (100) Flown 1984 RD-42 Aerojet MON/MMH 200 (91) 280 1 h 10 (4.5) Flown 1978 RD-40 Aerojet MON/MMH 900 (408) 295 Reusable on shuttle 15 (7) Flown 1977 RCS (indefinite) LMDE Grumman–Grumman, MON/A-50 10,000–1000 305–280 2000 s 300 (136) Flown on all Apollo 1967 (NGLT) (4536–454) lunar landing (formerly TRW) missions Delta 3910 second- NGST MON/A-50 9800 (4445) 308 2000 s 245 (111) Flown 1974

stage engine SACKHEIM RS-4-1 Rocketdyne MON/MMH 4000 (1814) 305 200 s 30 (14) Flight qualification 1988 Peace Keeper fourth stage axial Delta-V engine RL-60 Pratt & Whitney O2=H2 60,000 (27,216) 460 1000 s 700 (318) On hold Not complete – – – MB-XX Rocketdyne, MHI O2=H2 35,000 69,000 462 1000 s 600 1200 (272 544) In test Not complete (15,876–31,298) Electric thrusters – Resisto-jet NGST, Aerojet N2H4 NH3 0.1 (0.045) 280 300 100 h 4 (1.8) Flown 1980 – – – – – ARC-jet Aerojet N2H4 NH3, H2 0.05 1[1 25 kW] 550 900 100 h 5 25 (2.3 11) Flown 1986 ION thrusters NASA, Aerojet Xe 0.000001 3000 (3 kW) to 25,000 h 30 (14) Flown 1988 (0:5  107) 9000 (50 kW) HALL thrusters Russia, NASA, Xe 0.0001 (4:5  105) 1200 (0.5 kW) to 100 h 15 (6.8) Flown 1994 Aerojet, Busck 3500 (20 kW) Pulsed plasma Europe, Russia, Teflon 0.0000001 500 1  106 pulses 5 (2.3) Flown 2000 Aerojet (0:5  108) MPD NGST Lithium 10,000 (4536) 5000 1  106 pulses 30 (14) Preliminary Not flown development 6

Downloaded by Univ of Wisconsin-Madison on October 8, 2012 | http://arc.aiaa.org | DOI: 10.2514/1.23257 10.2514/1.23257 DOI: | | http://arc.aiaa.org 2012 8, on October Wisconsin-Madison of by Univ Downloaded fl PIT NGST Decomposed N2H4 0.5 at 300 kW 6,000 at 300 kW 1  10 pulses 50 (23) Preliminary Not own NH3, development Argon, Kr, Xe SACKHEIM 1315

Table 2 Causes of launch-vehicle failures

Launch-vehicle subsystem failures (1980–1999) Country Propulsion Avionics Separation Electrical Structural Other Unknown Total United States 15 4 8 1 1 1 30 CIS/Soviet Union 33 3 2 1 19 58 Europe 7 1 8 China 3 1 2 6 Japan 2 13 India 1 1 1 1 1 5

modified Apollo approach of more extensive hardware testing, but not act to shore up and sustain the rapidly diminishing, but still above guided by today’s more advanced modeling capabilities is highly “critical mass,” national rocket propulsion capabilities, it will soon recommended to minimize the potential for flight failures. be lost to the next generation of engineers that will subsequently be The development of advanced software simulation tools has entering into the propulsion community. A good example of this helped to reduce cost and development time by providing the engine problem was the major unplanned cost overruns (50–100%) design team with the capability to perform dynamic system analysis experienced on the recent RS-68 booster engine development before hardware fabrication. These tools provide valuable design program. insight and can lead to rapid system optimization during the early Returning to the big picture of the state of propulsion technology phases of a development program. This optimized system design today, it is important to note that many of the new booster and upper- approach should result in an engine design that is inherently more stage engines that were initiated by NASA and the U.S. Department reliable with a significantly lower development cost. The ability to of Defense (DoD) were stopped or reduced to the point of adequately model hardware and systems also speeds up the whole ineffectiveness because the commitment and funding required to development activity and generally tends to reduce the number of solve various complex problems that appeared, as they always do, “test–fail–fix and redesign” cycles. early in development and technology projects were lacking. This observation on the cost-driving characteristics of the Furthermore, underlying the basic lack of propulsion and space hardware applies to the recurring/operational costs for expendable transportation technology development activities in the last several launch vehicles as well. The average recurring cost breakdown for decades is the truly critical need to advance research in all areas of the Atlas, Centaur, Delta II, Titan II, and Titan IV launch vehicles is chemical (e.g., new higher-energy/higher-density propellants, 71% for the vehicle hardware. Of that hardware, the next level of combustion devices, advanced turbomachines, etc.), high-power breakdown shows that 54% of those costs are driven by propulsion electric and plasma (e.g., Lorentz force accelerators, etc.), and elements (engines, strap-on boosters, etc.), not including the tanks, advanced nuclear (e.g., fission, fusion, antimatter, etc.) propulsion which are classified as part of the vehicle structure for bookkeeping for the future exploration mission needs. purposes. So here again, a concerted effort to reduce engine costs will Even though a significant amount of money has been spent on also greatly reduce expendable launch vehicle recurring and launch-vehicle, spacecraft, propulsion-system, and engine develop- development costs. As a further cost-reduction observation, it can be ment programs since the completion of space shuttle development, readily shown that engine costs are mostly a function of operating most of these programs and engine projects were canceled early in chamber pressure (pc) and parts count, which basically is a strong the development cycle. The farthest progress that most chemical function of the type of power cycle (i.e., pressure-fed, gas generator, engine projects achieved was development testing and only one expander, or staged combustion). program (besides RS-68) proceeded past the protoqual stage. It should be readily apparent from the above key points that However, the expertise needed to support future human rating of simplified engine designs and existing databases, in combination rocket engines and propulsion systems comes primarily from direct with effective analysis/modeling tools, should go a long way toward involvement in development, qualification, and flight activities. reducing the development and recurring costs for booster and upper- Because no chemical engine (except RS-68) reached these stages stage engines for future expendable heavy-lift launchers, and in- since the shuttle began flying, it can be argued that the funding spent space and descent/ascent engines and systems for human spacecraft. during this time has not provided the needed experience to support However, it must also be cautioned that if the U.S. government does future space exploration. In fact, as the number of new space transportation and propulsion Downloaded by Univ of Wisconsin-Madison on October 8, 2012 | http://arc.aiaa.org DOI: 10.2514/1.23257 development programs have greatly diminished over the past two- plus decades, the response of the industrial base has been to consolidate and merge and to greatly shrink their combined capabilities and respective work forces. Some of the more recent and more significant consolidations are summarized in Figs. 5 and 6. These consolidations are already indicating an alarming decline in the previously dominant position of the U.S. aerospace industry globally. Further reductions could seriously inhibit the U.S. capability to compete in our domestic, as well as the international, marketplace.

IV. Summary of Lessons Learned from the Apollo and Space-Shuttle Programs The Apollo series of missions to the moon stands as one of the single greatest achievements in the history of space exploration. Two great super powers, the United States and the Soviet Union, entered into a race to be the first nation to land humans on the lunar surface and then return them safely to Earth. Both nations experienced Fig. 4 Approach for enabling well-thought-out, relevant, and balanced serious accidents resulting in crew fatalities during the early stages of engineering solutions. their respective moon landing programs. This tragic loss of life 1316 SACKHEIM

marred both nations’ human lunar landing activities and underscored the high degree of difficulty and risk associated with human space- exploration missions. Because of potential gains in global stature, however, both nations considered winning the high-profile geopolitical cold war race to be well worth the risks and cost, so the race to the moon continued beyond the impacts of the terrible experiences on both sides. Despite this dedication, the Soviets encountered increasingly difficult technical problems, resulting in significant failures of their moon rocket launcher, the N-vehicle, and additional loss of life. Only the United States was finally successful in landing two men on the moon and safely returning them to Earth, as directed by President John F. Kennedy’s national imperative. This unprecedented accomplishment was enabled by an extraordinary program/project management team, supported by an outstanding, highly talented engineering team drawn from all facets of the American aerospace community and backed by the technological and financial resources of the most powerful nation in the world. In the words of former Apollo Program Manager George E. Mueller, “Apollo was the first space system of systems ever created and implemented” [1]. It might now be worthwhile to look back at some of the specific achievements that were required to accomplish this remarkable, singularly successful series of events, in the interest of obtaining some lessons learned that can be applied to future exploration missions. This is especially relevant in light of the recognition that future space exploration by humans will require an extremely complex and interrelated space system-of-systems approach. From launch vehicles through a myriad of in-space vehicles, a complicated array of systems and infrastructure must be designed, developed, and deployed to enable successful completion of all of the missions required to ultimately land humans back on the moon and then on Mars. Fig. 5 Recent major consolidations of U.S. space and launch-vehicle As a further basis for using the lessons of the highly successful contractors. Apollo program, it must be remembered that there was actually a sequence of 10 successful individual Apollo missions—Apollo 8 through 17. Six of these actually enabled a total of 12 American astronauts to walk on the moon and return safely to Earth. Much was learned from each Apollo mission and applied to the subsequent one, culminating in the final lunar exploration and scientific research mission, the first and only human-based lunar geological survey, performed by the Apollo 17 crew. This is an important precedent for the complex exploration program that will be conducting all of the robotic and human missions planned for over a 30-plus year period and culminating with multiple safe and sustainable landings of humans on the planet Mars for relatively long-duration stays on the surface of the planet. At the time, Apollo was considered to be the most ambitious engineering project ever undertaken, a task that was considered to be far more complicated than the Panama Canal and the Manhattan

Downloaded by Univ of Wisconsin-Madison on October 8, 2012 | http://arc.aiaa.org DOI: 10.2514/1.23257 Project combined. The idea of designing and building a 36-story- high rocket, loading it with more equivalent energy than an atomic bomb, putting three men on top, launching them to the moon, safely landing, and then returning them to Earth seemed even to some of the excellent engineers of that day to be just about impossible, Mercury- Redstone had just completed its initial suborbital flight in May 1961, and would require nothing short of a miracle to complete. Nonetheless, the task was accomplished multiple times with higher reliability than any rocket in history. The Apollo spacecraft alone required 500  106 man-hours of work. It contained more than 2  106 functioning, separate parts that had to survive intense launch and boost vibrations and loads, then function in the weightless void of space, where the temperature range between sunlight and shadow was greater than 600F (315C). By far, the most critical technologies that had to be designed, developed, and matured were the rocket propulsion elements required for all of the critical and serial operating modes of the numerous space transportation systems in this unprecedented system of systems. Ten new rocket engines had to be designed, qualified, and human rated from scratch (Table 3); no engine existed anywhere in the world that could operate at the desired design conditions or over the required Fig. 6 Recent mergers of U.S. propulsion suppliers. ranges [e.g., 1:5  106-lbf (680,389-kgf) thrust for each of five main SACKHEIM 1317

Table 3 Number of rocket engines developed for the Apollo mission

Saturn 5/F-1 engine F-1 S-2 second-stage S-1VB J-2 RCS thruster R4-B Solid escape tower rocket ER SE-7 ullage control S-2 SE-7 SE-8 command module RCS S-14B SE-8 TRW ullage rocket UC-1 Service module main engine AJ10-137 Lunar descent engine LMDE Lunar ascent engine LMAE Total no. of new engines developed for Apollo 10

Fig. 7 NASA space transportation budget 1959–2000. booster engines and 10:1 deep throttling for the lunar module descent stage to land on the moon]. This was the all-time peak activity for the then-fledgling U.S. as part of the vehicle structure. Therefore, a significant portion of the rocket industry, as shown in Figs. 1 and 2. The United States was cost of vision for space exploration will be the design, development, rapidly developing multiple liquid-stage and in-space propulsion certification, and manufacture of propulsion hardware and systems. systems required for the Mercury, Gemini, and Apollo programs. Of the many incredible development successes leading up to the One of the primary factors contributing to the successful first landing of humans on the moon, the operational success of all the development of these systems was the knowledge and technical propulsion elements and the overall demonstrated flight launch expertise gained during the liquid propulsion engine development vehicle and propulsion-system reliability were truly awesome. As efforts required for the Intercontinental Ballistic Missile (ICBM) shown in Table 4, every time a Saturn V rocket lifted off for the programs (i.e., Redstone, Jupiter, Juno, Thor, ATLAS, and TITAN I moon, 86 rockets/thrusters had to work as designed for complete and II). The ability to leverage experience gained during these mission success. In many cases, because of weight, volume, and “real programs accelerated the development of liquid propulsion-system estate” access and constraints, many of these rocket engines and/or expertise in the United States. The explicit benefits of this early liquid thrusters could not be redundant (Table 4). ICBM engine development experience soon became apparent in the Furthermore, as shown in Table 5, at least 19 serial discrete early stages of mission design during the 1960s, because it was soon dynamic events (e.g., staging, rendezvous, landing, ascent, etc.) had determined that three different propellant combinations would be to occur flawlessly for complete mission safety and success. It is also required to meet the launch-vehicle gross liftoff weight (GLOW) remarkable to note that for the 30 or so launches of humans during the constraints and the overall performance, reliability, and mission Mercury, Gemini, and Apollo programs (all aimed at enabling the fi constraints. For the Saturn V rst stage, LO2=kerosene was selected United States to land the first humans on the moon) the human-rated to ensure the highest propellant mass fraction and the minimum launch vehicles (i.e., Redstone, Atlas, Titan II, and Saturn V) structural mass. For the second and third stages, as well as during the demonstrated a flight success record and reliability of 100%, again translunar injection (TLI) burns, LO2=LH2, which had never been unprecedented for that period. flown before, was selected to provide the highest possible chemical It was only through exceptional engineering, scientific, performance. This provided the highest possible in-space payload mechanical and electrical skills, innovation, and dedication that fractions at that time. Spontaneous ignition using hypergolic this record of success could have been achieved. An awesome – N2O4=50 50 unsymmetrical dimethylhydrazine (UDMH) N2H4 government, industry, and academic team was assembled that yielded the reliability and multiple-start in-space capability required included some of the greatest technical minds in the country for guaranteed in-space engine start and restart, multiple-pulsing including Werner Von Braun and his team of German rocket reaction/attitude control, Earth–moon cruise and return for the scientists from Peenemunde, Joe Shea from MIT, Brainard Holmes, service module, lunar descent, ascent back to lunar orbit, and return James Webb, T. Keith Glenan, Maxime Faget, Harrison Storms, Joe to Earth. Gavin, Leland Atwood, and many thousands more who were chosen Six different rocket engine companies, two of which are long out and trained to “grow up” technically in this highly innovative period of business, and five different prime contractors, most of which are of technical achievement. either out of aerospace or completely out of business today, Many of these people drove themselves beyond normal limits of

Downloaded by Univ of Wisconsin-Madison on October 8, 2012 | http://arc.aiaa.org DOI: 10.2514/1.23257 collaborated under contract to NASA to enable the very rapid and human endurance to ensure the success of this colossal endeavor, highly successful development of the 10 new rocket engines, ranging which was a testimony to their dedication to the nation’s exploration from 1:5  106 lbf (680,389 kgf) of thrust down to 100 lbf goal. It is clear that we will need to find such people again to enable (45 kgf) of thrust (for the RCS thrusters for the Command, Service, the success of the exploration missions planned for the future. This and Lunar modules) and the associated vehicle main and auxiliary will be difficult due to the decline of the American aerospace base propulsion systems for Apollo. Approximately 80% of the $75 and the diminution of the national technical community’s billion (in today’s dollars) spent for the design, development, and capabilities. human rating of the entire fleet of Apollo/Saturn systems and As for the human rating of space launch vehicles and flight hardware was spent on propulsion and associated elements including systems, some additional experienced-based observations are in the test infrastructure to develop and certify systems. Figure 7 shows order here. It was concluded quickly in the early phases of the Apollo the dollars spent by NASA for the development of space RDTE efforts that more than just analysis or probabilistic/statistical transportation vehicles since that time for comparison purposes. analytical risk assessment would be needed to guarantee the required Also, for comparison purposes, the overall cost of all of the level of mission success, reliability, and confidence in the design propulsion elements for all of the space-shuttle propulsion elements approach or meet the overall safety goals that were required to send was about 75% of the complete shuttle program from both a research, humans to the moon and return them safely to Earth. The name of the development, test, and evaluation (RDTE) and an operational point reliability/mission confidence-building game had to be test, test, test, of view. Examination of historical rocket and jet engine development and then more testing. has shown that roughly 50% of the development cost is hardware In fact, it also became apparent that there are really five basic related (the rest evenly split between labor, test, engineering, and approaches to “human rating” a system of systems and all of its management). For expendable launch vehicles it was found that integral elements. These are as follows: 54% of the hardware recurring costs were driven by propulsion 1) by selecting a sufficiently large number of specimens of the new elements and this did not include tanks because they were classified product design to be tested; 1318 SACKHEIM

Table 4 Total number of rocket engines/motors per flight of each of Apollo 11–17 missions

 6 — S-IC stage F-1 Engines, LO2=RP-1=1:5 10 lbf Rocketdyne (R/D) 5 Retrorockets, solid fuel-Thiokol 8 S-II stage Ullage rockets, S-IC/S-II interstage 100 lbf—R/D 4 Retrorockets for two auxiliary propulsion-system modules solid rockets—Thiokol 4 — J-2 engines LO2=LH2=200; 000 lbf R/D 5 S-IVB stage Main ullage rockets, jettisonable 2 Rockets for two auxiliary propulsion systems modules 100 lbf—TRW 8 Fwd. compartment reaction control engines (pitch) 100 lbf—Marquardt 2 J-2 engine 1 Command module Aft compartment reaction control engine (pitch) 100 lbf—Marquardt 10 Roll engines 100 lbf MARQUARDT 2 — Service module Service module engine (SME) N2O4=A-50, 20,000 lbf Aerojet 1 RCS engines 100 lbf 16 – — Descent engine, N2O4=A-50 10,000 1000 lbf deep throttling TRW 1 Lunar module Ascent engine, N2O4 3000 lbf Bell/R/D 1 RCS engines 100 lbf—Marquardt 16 Total 86

2) by conducting a sufficient number of tests on each prototype and limits of their thermal, structural, and dynamic capabilities. The way flightlike specimen as well as robust subsystem and system-level to resolve these dangerous issues for new rocket propulsion systems tests to demonstrate adequate operating and performance capabilities is exactly as was implemented for the Apollo program. Find and under nominal operational and environmental conditions; employ the most knowledgeable, experienced, and brightest 3) by conducting significant off-nominal and/or overstress propulsion engineers available at the beginning of the new program conditions for testing to demonstrate adequate new product design and in the early design and development phases, and then continue to and system margins; bring new, very smart, but still inexperienced engineers onboard so 4) to design the new product systems with as much redundancy as they can be trained and eventually provide equally valuable feasible to minimize single-point failures; and finally contributions and make good technical decisions before the program 5) to use a logical optimized combination of all of the above enters the flight phases. approaches to establish sufficient confidence in the human-rated An example of the benefits of using the above approach for capabilities, reliability, and safety of the new products and their designing and certifying high-reliability human-rated flight systems associated integrated systems under all anticipated operational can, perhaps, best be illustrated by examining the reliability of the conditions. world’s launch vehicles. The launch and boost phase of a space flight This was the methodology employed to achieve the confidence in program is considered to be the most dangerous phase of any space the human-rated capabilities of all the elements and systems mission, because of the huge amount of propulsion power that must associated with the Apollo program and was certainly a major factor be generated, controlled, and released. The launch vehicle must add in achieving such a stunning series of mission successes and an 9100 m=s to take a payload from Cape Canaveral/KSC, Florida to incomparable record of mission reliability and safety. The human- a minimum stable Earth-parking orbit. rating approach used during Apollo, and not strict reliance on two- With this in mind then, it is very interesting to review the launch fault tolerance, was the paramount approach to achieving the vehicle/phase success rate for all space-faring nations for about the reliability and performance for the propulsion and space last 30 years (1970–1999). There have been about 3450 launches, transportation elements of Apollo. both manned and unmanned, during this period. The success rate for Because of the need to save weight on flying space vehicles by all these launches combined has remained essentially flat at about 93 maximizing propulsion performance, propellant/propulsion hard- to 94%. If one were to look just at the human space launches ware mass is typically 80–90% of the takeoff GLOW for any worldwide during this period, a very different conclusion emerges. mission, the propulsion engines and all other elements have to To date, the world has launched about 242 human space flight operate at maximum performance and at the maximum acceptable missions with only one launch failure of a reusable launch vehicle (i.e., STS-51L or Space Shuttle Challenger) being reported. This

Downloaded by Univ of Wisconsin-Madison on October 8, 2012 | http://arc.aiaa.org DOI: 10.2514/1.23257 results in a human space launch phase success rate of 99.6%, or very Table 5 Number of staging/dynamic events on a nominal Apollo close to the minimum value of 99.8% sought by Von Braun’s mission Saturn V rocket team essentially broken down as follows: First stage 1 1) about 242 total manned space launches worldwide (U.S., Second stage 1 USSR/Russia, China); Third stage 1 2) 241 successful launches (99.6%); Escape tower, jettison 1 3) STS51L (Challenger) lost; CSM to lunar orbit after finishing TLI 1 4) 238 successful landings (98.8%); Interstage separation 1 5) STS 107 (Columbia) lost on reentry; CSM to LEM docking 1 6) Soyuz-1 lost on landing (tangled chutes); LEM removal from launch shroud 7) Soyuz-11 crew lost on reentry (asphyxiated); LM rotation/reconnect 1 fl LM separation from CSM in lunar orbit 1 8) total manned ight success rate is about 98.3%; Lunar landing 1 9) worldwide total manned and unmanned launch success rate LM egress and surface activities 1 from 1970 to 1999 (about 3449 launches) essentially unchanged at Return to LM ascent stage blastoff 1 about 94%; Rendezvous with CSM 1 10) a challenge is to increase manned flight safety given a 30-year Leave lunar orbit 1 history of unmanned launch success rates that are essentially flat. Enter TEI trajectory 1 This comparison of results over the last 30 years of the 20th Stage off service module 1 century (the dawn of the Space Age) shows how much can be Earth reentry and kinetic braking 1 achieved by very smart propulsion engineers and managers, Deploy parachutes 1 fi Total no. of events 19 combined with a development and certi cation methodology of stringent repeated testing, and strongly reinforced by a mindset that SACKHEIM 1319

demands that the development team test as they plan to fly and fly failures, and acquisition costs of payloads and expendable vehicles. fl fi only as they have tested. For example, referring to Table 2, it can be The in uence of delivered speci c impulse (Isp) on overall mission seen that for the last 20 years (1980–1999), 64 out of a reported costs per year is very small, particularly for the first stage. 114 failures, which equates to about 56%, were caused by There are technologies in development that may permit cost- propulsion-system failures for all reported launches, worldwide. The effective upgrades to various system elements of Atlas V and next highest cause of vehicle failure was a tie between avionics and Delta IV, such as upper stages. Most of these technology-based separation/staging subsystems at about 10%. Refer to the numbered upgrades would be justified based on eliminating critical failure list above and Table 2 for more detailed information and data modes or increasing margins against known retained failure modes. on launch vehicles and associated subsystem failure rates and causes. If one were to examine the development and flight history of the U. A. Large Launch Vehicles S. space shuttle or STS program, very similar lessons about the 1. Delta IV Family importance of good propulsion engineering and knowledge, The Boeing Delta IV family of two-stage launch vehicles uses a combined with a comprehensive, well-thought-out development and common 5-m diameter first stage powered by a single P&W/R/D flight certification program would also be apparent for the future (Pratt and Whitney Rocketdyne) rocket engine (RS-68) operating on missions. The space shuttle was a very different type of human space LO2 and LH2. The baseline vehicle, designated as Delta IV Medium, flight program from Apollo. In some ways it was more difficult due to has a 4-m diameter second stage powered by a Pratt and Whitney RL- fl fi the reusability requirements and goals (100 ights) established by the 10B-2 engine using LO2 and LH2. Three other con gurations of program (i.e., it was the first and only reusable launch and astronaut Delta IV Medium vehicles offering progressively more payload return space transportation system). These requirements were weight to LEO or geosynchronous transfer orbit (GTO) are in the extremely difficult and completely changed the concept of rocket family. These three vehicles use Alliant Techsystems GEM-60 (60- engines, thrusters, and main propulsion systems. in diam) graphite-epoxy solid-propellant motors as strap-on In other ways the space shuttle system could be considered far less boosters. challenging than Apollo. The shuttle only had to carry humans to As shown in Fig. 8, the vehicles are designated as Medium+ (4,2); LEO, a distance of 200–400 mi (322–644 km), and no further, Medium+ (5,2); and Medium+ (5,4). The first numbers indicate the whereas Apollo had to travel about 240,000 mi (386,243 km) to the diameter of the second stage and payload fairing, and the second moon, land safely, and return with the human cargo, even though it numbers designate the number of graphite-epoxy motor (GEM) was an expendable space flight vehicle. In addition, Apollo was strap-ons. All of the 5-m-diam second stages also use longer extremely weight constrained, much more than the space shuttle propellant tanks. The fifth vehicle in the family is called Delta IV human/cargo-carrying STS, which increased the technical challenge. Heavy. The configuration uses three of the common 5-m-diam first Here again, the overall cost of the propulsion engines, thrusters, stages in parallel. The second stage uses the same 5-m diam, longer components, and main and auxiliary propulsion systems proved to be tank that is used on the Medium + (5,2) and (5,4) vehicles. much higher than anything else in the system, both for development This family of vehicles is stated to have the capability to deliver and flight certification, as well as for flight hardware operation and from 20,000 to 48,000 lb (9072–21,772 kg) to LEO or 9300 lb to maintenance. 28,000 lb (4218–12,700 kg) to GTO. The propulsion-system The overall cost of all of the propulsion elements for all of the elements that essentially control the various configuration’s space shuttle propulsion elements was about 75% of the complete performance and risks are the first- and second-stage engines and shuttle program from both an RDTE and an operational point of the solid-propellant strap-on motors. These are described below. view. So, as always, the propulsion elements for a major STS, space flight vehicle, or space system of systems are the most difficult to develop, initially the most unreliable, sometimes well into the flight program, and the most expensive hardware throughout all phases of any space system flight program.

V. Current Status of Space Propulsion Technology In Sec. II we saw the degree to which the success of the Apollo program was tied to successful developments in space propulsion, and in Sec. III, we saw that the nation’s current capability in space

Downloaded by Univ of Wisconsin-Madison on October 8, 2012 | http://arc.aiaa.org DOI: 10.2514/1.23257 propulsion is but a shadow of the capability put in place by the ICBM and Apollo programs. A more penetrating look at this capability will illustrate that on which the exploration program can depend today. At the present time, with the exception of NASA’s stated intent to develop a shuttle-derived series of launch vehicles for the exploration mission, there are no plans to replace the Atlas V or the Delta IV launch vehicles until sometime beyond 2020. This is a realistic observation because it is highly unlikely that any technologies currently in development or envisioned for liquid- or solid-propellant all-rocket (noncombined-cycle) propulsion systems for space access would demonstrate the levels of performance improvement or reduction in operational risks and costs necessary to justify the huge cost of an Atlas V/Delta IV or even larger class new- centerline launch-vehicle development. This includes technologies focused on reusable vehicles and propulsion systems. The very large range of payload weights deliverable to a broad spectrum of orbits made possible by the modularity configurations available for both Atlas V and Delta IV families of vehicles also makes it unnecessary to develop new launch vehicles having slightly higher performance for any given configuration. The yearly costs for these low-intensity, limited flights per year are driven by site maintenance, cadres of trained personnel and managers, realized Fig. 8 Delta IV family of launch vehicles. 1320 SACKHEIM

Table 6 RS-68 engine characteristics Table 7 RL-l0B-2 engine characteristics

RS-68 Characteristics Details Thrust level 100% 60% Thrust 24,750 lbf (11,226 kgf)) Thrust, vacuum 745 lbf (338 kgf) 440 lbf (200 kgf) Weight 664 lb (301 kg) —— Weight 14,560 lbf (6604 kgf) Fuel/oxidizer LO2=LH2 Thrust, s/l 650 lbf (295 kgf) 345 lbf (156 kgf) Mixture ratio 5:88:1 Engine mixture ratio 6 6 Isp, vacuum 465.5 s Isp, vacuum 410 s 410 s Starts (total) 15 Isp, sea level 365 s 365 s Service life (total) 3500 s Chamber pressure 1410 psia (9722 kPa) 836 psia (5764 kPa) Expansion ratio 280:1 Expansion ratio (E) 21.5 —— Length (stowed)\ 86.5 in (219.7 cm) Length (deployed) 163.5 in (415.3 cm) Diameter (nozzle extension) 84.5 in (214.6 cm) a. Common Booster Core: RS-68 Engine. In the early 1990s, Rocketdyne initiated development of the first new U.S. domestic summarizes the GEM-60 vectorable nozzle operation and booster class engine in more than 25 years. This engine, designated performance characteristics. as RS-68, uses LO2=LH2 propellants and develops 650,000 lbf (294,835 kgf) of sea-level thrust. It is capable of being throttled to 60% of full power level. 2. Atlas V Family The RS-68 engine uses a simple, open gas-generator cycle with a The Lockheed Martin Atlas V family of two-stage launch regeneratively cooled main chamber. It was designed, developed, vehicles, shown in Fig. 9, uses a 12.5-ft (3.8-m) diam common-core and certified in a little over five years, and flew on the first Delta IV first stage. The first stage uses a single rocket engine (RD-180) launch in late 2002. The engine has far fewer parts than the SSME, operating on LO2 and kerosene (RP-1). The baseline vehicle which greatly contributes to its reduced production costs. Only 11 designated as the Atlas V 401 has a 10-ft (3.05-m) diam, 42-ft (12.7- major components are included: the main combustion chamber m) long Centaur second stage powered by a RL-10A-4-2 engine. The (MCC), single oxygen and hydrogen turbopumps, gimbal bearing, 400 series has a 13-ft (4-m) diam payload fairing, and the 500 series injector, gas generator, heat exchangers, and fuel exhaust duct. This provides a 16.5-ft (5-m) fairing. The Centaur stage can use either one amounts to an over 80% reduction of parts and a 92% reduction in or two RL-10A-4-2 engines. Depending on the mission, the 500 hand touched labor from SSME. The development cycle time was series can be configured with from zero to five strap-on solid rocket also much reduced, and the nonrecurring costs were claimed to have motors (Aerojet, Sacramento). Each motor provides 254; 000 lbf been reduced by a factor of 5 over previous cryogenic engines. The (1,115,212 kgf) of thrust at liftoff. The 500 series of vehicles has a engine performance and operating characteristics are summarized in capability to deliver from 20,000 to 45,000 lb (9072–20,411 kg) to Table 6. LEO (27.8) or from 8750 to 19,100 lb (3970–8664 kg) to GTO. The b. Second Stage: RL-10B-2, RL-10 Engine. The RL-10 engine propulsion-system elements are described below. ’ fi was the world s rst LO2=LH2 rocket engine operated in space. The a. Common First Stage: RD-180 Engine. The RD-180 is a two engine weighed 289 lb (131 kg) and developed an Isp of 410 s. Since thrust-chamber version of the original Russian RD-170, which is the first successful launch of an Atlas/Centaur RL-10 in 1961, nine used to power the first stage of the Zenit launch vehicle. The engine additional and different models of the RL-10 engine family have features an oxidizer-rich, staged-combustion cycle (ORSC) and been developed. burns LO2 and RP-1 propellants. It has a health monitoring and The RL-10B-2 is the latest derivative of the RL-10 engine which life prediction system. The two thrust chambers can gimbal 8 deg. features a carbon–carbon extendible nozzle. This high-expansion Minimal interfaces are used between the launch pad and the nozzle (285:1) enables the engine to operate nominally with a vehicle. chamber pressure of 633 psi (4364 kPa) and develops an Isp of The engine offers relatively clean operations with oxidizer start 465.5 s. This engine can lift payloads up to 30,000 lb (13,600 kg) and and shutdown modes that eliminate coking and the potential for currently powers the upper stage of the medium- and heavy-lift unburned kerosene pollution. Forty to 100% continuous throttling configurations of Boeing’s Delta IV launch vehicle in addition to the provides the capability for real-time trajectory matching and engine upper stage of the Delta III. The engine performance and operating checkout on the pad before launch commit. Key performance characteristics are summarized in Table 7. The full family of flight- and operating characteristics of the engine are summarized in certified RL-10XX engines is listed in Table 8 along with their Table 10. Downloaded by Univ of Wisconsin-Madison on October 8, 2012 | http://arc.aiaa.org DOI: 10.2514/1.23257 respective key design features. b. Second Stage: RL-10A-4-2 Engine. The RL-10A-4-2 is a c. Strap-On Solid Rocket Booster: GEM. The 60-in. (1.5-m) LO2=H2 closed expander-cycle engine. It is equipped with a single diam GEM motor is a strap-on booster system developed to increase turbine and gearbox that drives the LO2 and H2 pumps. Additionally, the payload-to-orbit capability of the Delta IV M+ launch vehicles. the engine features a dual direct spark ignition (DDSI) system and Two and four strap-on motor configurations of the GEM-60 can be can be flown with a fixed or extendible drop-down nozzle skirt. The flown on the Delta IV M+ vehicles. The motor features a ‡5 deg engine operates nominally with a chamber pressure of 610 psi canted, moveable nozzle assembly. This motor is a third-generation (4206 kPa) and develops an Isp of 451 s. A summary of the engine GEM with both fixed and vectorable nozzle configurations. Table 9 performance and operating characteristics is presented in Table 11.

Table 8 RL-10 engine model comparison

Model no. A-1 A-3 A-3-1 A-3-3 A-3-3A A-4 A-5 A-4-1 B-2 Vacuum thrust, 15,000 15,000 15,000 15,000 16,500 20,800 14,560 22,300 24,750 lbf (kgf) (6,804) (6804) (6804) (6804) (7484) (9435) (6604) (10,115) (11,226) Chamber pressure, 300 (2068) 300 (2068) 300 (2068) 395 (2723) 475 (3275) 578 (3985) 485 (3344) 610 (4206) 644 (4440) psia, kPa Thrust/weight 50 50 50 50 54 67 61 Expansion ratio 40:1 40:1 40:1 57:1 61:1 84:1 4:3:1 84:1 285:1 Isp, s 422 427 431 442 444 449 368 451 466.5 Flight certification November June 1962 September October November December August February May 1998 date 1961 1964 1966 1981 1990 1992 1994 SACKHEIM 1321

Table 9 GEM-60 motor with vectorable nozzle

Motor dimensions Weights, lbm (kgm) Motor diameter, in. (cm) 60 (152) Total loaded 74,158 (33,638) Motor length, in. (cm) 518 (1316) Propellant 65,471 (29,697) Motor performance [73 F (22.8 C) nominal] Case 3578 (1623) Burn time, s 90.8 Nozzle 2187 (988) Average chamber pressure, psia (kPa) 818 (5640) Other 2922 (1326) Total impulse, lbf-s (kgf-s) 17:95  106 (8:16  106) Burnout 8346 (3786) Burn time average thrust, lbf 197,539 Temperature limits Nozzle Operation 30–100 F(1:1–37:8C) Housing material 4340 steel Propellant designation Exit diameter, in. (cm) 43.12 (109.52) QEY, 87% solids HTPB Expansion ratio, average 11 Production status Production

c. Strap-On Solid Rocket Booster. The solid rocket strap-on 60 fps (18 m=s) while the first stage arms and prepares for ignition. booster motors have been developed, flight qualified, and produced The forward velocity of Pegasus during the descent is the same as the by Aerojet, Sacramento. The Atlas V family of launch vehicles uses launch aircraft, or Mach 0.8 [524 mph (843 kph)]. After 5 s in free from one to five strap-on solid rocket motors depending upon the fall, stage one’s solid rocket motor, manufactured by ATK, fires and mission and launch trajectory requirements. The solid rocket motors burns for 71 s. The Pegasus 22-ft (7-m), delta-shaped wing begins are ignited at liftoff and burn for over 90 s, each providing a thrust in to produce lift as the Pegasus accelerates, and the launch vehicle excess of 250,000 lbf (113,398 kgf). The motor key design features begins a 2.5-g pull-up. As Pegasus climbs, the booster experiences a are summarized in Table 12. maximum dynamic pressure (max-q)of1200 lb=ft2 psi (8274 kPa) approximately 30 s after first-stage ignition. [For comparison, on a typical space shuttle launch, max-q is equal to B. Small- and Medium-Sized Launch Vehicles approximately 600–700 lb=ft2 psi (4137–4826 kPa).] 1. Pegasus The second-stage Alliant Techsystems (ATK) solid-fuel motor The Pegasus is an air-launched (via a modified Lockheed L-101 I ignites 95 s into the flight at an altitude of 37 mile (60 km), and at aircraft), three-stage, all solid-propellant, three-axis-stabilized 2 min, the payload fairing is ejected. The second stage flies to an vehicle. It is manufactured by the Orbital Sciences Corporation. altitude of 129 mile (208 km) with a velocity of >12; 000 mph The Pegasus-XL vehicle, a “stretched” version of the original (19,312 kph). At the appropriate altitude to achieve the designated Pegasus vehicle, can place a 400–1000-lb (181–454-kg) payload into orbit, the third stage ATK motor ignites and burns for 1 min 6 s to LEO. The original, or standard, version of the Pegasus was retired in place its payload into orbit. 2000, and only the Pegasus-XL is used today. During a typical flight, the launch aircraft climbs to an altitude of 38,000 ft (11,582 m) and 2. Athena the Pegasus-XL is released from the belly of the L-101 1. The The Athena program began in 1993 under the sponsorship of Pegasus-XL begins an unpowered descent at a rate of approximately LMSC, Sunnyvale, California. The first operational mission of the Athena, Athena I, successfully launched the NASA Lewis satellite of 1750 lb (744 kg) mass into orbit on 22 August 1997. The first Athena II was successfully launched on 6 January 1998, sending NASA’s Lunar Prospector spacecraft of 4350 lb (1896 kg) mass on its mission to study the moon. The Athena propulsion specifics are given in Table 13, although it is believed that this vehicle is no longer available or in production.

Table 10 RD-180 engine characteristics

Nominal thrust Sea level 860,200 lbf (390,178 kgf) Downloaded by Univ of Wisconsin-Madison on October 8, 2012 | http://arc.aiaa.org DOI: 10.2514/1.23257 Vacuum 933,400 lbf (423,383 kgf) Isp (sea level) 311.3 s Vacuum Isp 337.8 s Chamber pressure 3722 psia (25,662 kPa) Nozzle area ratio 36:87:1 Mixture ratio 2.72 Length 140 in. (356 cm) Diameter 118 in. (300 cm) Throttle range 40–100 % Total system dry weight 11,889 lb (5393 kg)

Table 11 RL-10A-4-2 engine characteristics

Weight with nozzle 386 lb (175 kg) Length (approximately) 91.5 in. (232 cm) Nozzle extension 20 in. (51 cm) Nozzle area ratio 84:1 Isp vacuum, s 451 Thrust 22,300 lbf (10,115 kgf) Propellants LO2=LH2 Nominal mixture ratio 5:5:1 O/F Fig. 9 Atlas V family of launch vehicles. 1322 SACKHEIM

Table 12 Atlas V strap-on rocket motor design features

Designed for maximum payload flexibility 1–5 interchangeable solid rocket motors SRM size optimized for vehicle design 62 in. (1.57 m) diam, 64 ft (19.5 m) total length Designed for reliability/robustness Fixed nozzles Monolithic—no. segment joints High structural and thermal margins Designed for producibility Heritage processes and materials Lean production processes, tools, and facilities Designed concurrently with airborne hardware Designed for operability Fully integrated at the manufacturing site and shipped ready to fly

Table 13 Propulsion specifics of Athena launch vehicle

CASTOR 120 ORBUS 21D Stages Athena I First Second Athena II First and second, composite case Third, composite case Thrust vector control Blowdown cold gas-powered hydraulic Electromechanical Nozzle Carbon phenolic Carbon phenolic Length, in. (m) 347 (9) 124 (3) Diameter, in. (m) 92 (2) 92 (2) Thrust, lb (kg) 435,000 (197,313) 43,723 (19,832) Propellant Class 1.3 AP/HTPB Class 1.3 AP/HTPB Contractor Thiokol Pratt & Whitney

3. Taurus C. Integrated High-Payoff Rocket Propulsion Technology Activities Taurus is an extended-capability, ground-launched version of the The integrated high-payoff rocket propulsion technology Pegasus launch vehicle. It uses three stages of the Pegasus boosted by (IHPRPT) program was initiated under the Air Force Research a large Castor (120) solid-propellant motor. It is designed to launch Laboratory (AFRL) sponsorship in 1994. It is a joint government and small and Med-Lite satellites up to 2976 lb (1350 kg) into LEO. industry effort focused on affordable technologies for revolutionary, Liftoff weight varies between 150,000 and 220,000 lb (68,039 and reusable, and rapid-response military global reach capability. It 99,790 kg). Four variants of the Taurus launch vehicle exist. The addresses sustainable strategic missiles, long life or increased smallest version, known as the ARPA Taurus, uses a Peace Keeper maneuverability, spacecraft capability, launch-vehicle propulsion, first stage instead of a Castor 120 motor. A second size uses the C 120 and high-performance tactical missile capability. IHPRPT attempts first stage and a slightly larger Orion 50S-G second stage. The to emulate the IHPTET aircraft engine program, which has been very Taurus XL uses the Pegasus-XL rocket motors (Orion 50S-XL and successful in the development and testing of new turbine jet engine Orion 50-XL) and is considered to be an improved launch-vehicle technologies. version. The largest Taurus variant, the Taurus XLS, is a study-phase Although funding for the program has been limited, contractors vehicle that adds two Castor IVB solid rocket boosters to the have had considerable freedom to develop new technologies with the Taurus XL to improve payload lift capability by 40% over the potential to improve the performance and life of both solid and liquid standard Taurus. For all Taurus configurations, satellite delivery to a rocket engines as well as for in-space propulsion technologies. The GTO orbit can be achieved with the addition of a Star 37 FM perigee major difficulty voiced by several of the contractors is that there is no kick motor. clear definition of near-term and future USAF propulsion needs. Specific performance goals have been established for each element of the program, some of which are summarized in Table 14. “ ’’ 4. Minotaur The building block approach to planning for technology insertion seems to be working well and furthermore there appears to Minotaur is a low-cost, four-stage launch vehicle using a be excellent cooperation between the USAF and contractors and also Downloaded by Univ of Wisconsin-Madison on October 8, 2012 | http://arc.aiaa.org DOI: 10.2514/1.23257 combination of Minuteman II motors and other proven space launch between the USAF and NASA. It provides a basis for government technologies. The Minuteman rocket motors serve as the vehicle’s and industry to walk in and step on an evolving but common path. first and second stages, efficiently reusing motors that have been Under the IHPRPT program, a new high-performance 40,000-lbf decommissioned for military applications, as a result of arms (18,144-kgf) rated LH turbopump, sponsored by AFRL, is currently reduction treaties. Minotaur’s third and fourth stages, structures, and 2 under study and may be built and tested to validate the improved suite payload fairing are common with the Pegasus-XL rocket. Its of design/analysis software. Originally, there was also to be an capabilities have been enhanced by OSC, with the addition of advanced regen-cooled thrust chamber assembly built for a similar improved avionics systems. purpose, but this was struck from the program due to funding The primary program goal is to provide low-cost, reliable space limitations. The only detail design work on the program is directed to launch capability in support of U.S. government small-satellite the LH turbopump assembly (TPA), and the only testing on the launch requirements. The program was structured to retire most of 2 program will be on the LH TPA. the development risk with the first mission launch followed by 2 routine launch services operations. The Minotaur is considered a small launch vehicle. It can lift 750 lb (340 kg) to a 400-nmi (741- km), sun-synchronous orbit, which is 1:5 times the Pegasus-XL VI. Future Small Launcher Program: FALCON capability. The Minotaur can operate with either of two fairings, The FALCON program is funded by the Defense Advanced allowing for the launch of oversized payloads when required. The Research Projects Agency (DARPA), with USAF and NASA standard configuration uses a slightly modified Pegasus fairing. It has support. After the initial 6-month phase I study, four contractors were a dynamic payload volume of about 46 in. (1.2 m) (diameter) by selected for phase IIb. These were 1) Air Launch LLC; 2) Lockheed 88 in. (2.3 m) (length). The larger fairing has a dynamic payload Martin Corp.; 3) Microcosm Inc.; and 4) Space Exploration volume of about 61 in. (1.5 m) (diameter) by 133 in. (3.4 m) Technologies. Phase II has been divided into three parts, phase IIa (length). covers from phase II authority to proceed (ATP) to preliminary SACKHEIM 1323

Table 14 IHPRPT phase I, II, and III goals (2000–2010)

IHPRPT goals 2000 2005 2010 Boost and orbit transfer propulsion Reduce stage failure rate 25% 50% 75% Improve mass fraction (solids) 15% 25% 35% Improve Isp, s (liquids) 14 21 26 Improve Isp, s (solids) 5 10 20 Reduce hardware costs 15% 25% 35% Reduce support costs 15% 25% 35% Improve thrust to weight (liquids) 30% 60% 100% Mean time between removal (mission life reusable) 20 40 100 Spacecraft propulsion

Improve Itot=mass wet† (electrostatic/electromagnetic) 20%=200% 35%=500% 75%=1250% Improve Isp (bipropellant/solar thermal) 5%=10% 10%=15% 20%=20% Improve density Isp (monopropellant) 30% 50% 70% Improve mass fraction (solar thermal) 15% 25% 35% Tactical propulsion Improve delivered energy 3% 7% 15% Improve mass fraction (without TVC/throttling) 2% 5% 10% Improve mass fraction (with TVC/throttling) 10% 20% 30%

design review (PDR). Phase IIb covers from PDR to critical design contractors developed several technologies to meet these goals. review (CDR), and phase IIc goes from CDR to flight demonstration Some of the key technologies developed or optimized that could be of the Smallsat Spacelift mission [2]. modified for or transferred to other programs are ablative thrust The goals for phase IIa were to develop the vehicle design to a chambers, self-pressurization systems (VAPAK, Tridyne, etc.), low- PDR level. Concept of operations (CONOPS) was also to be detailed cost avionics, hybrid combustion (using a patented staged- in this phase. Technical risk-reduction activities were defined and combustion concept), and composite tanks. Ablative thrust chambers executed. The teams interfaced with the launch ranges and then were used by at least two contractors as an alternative to actively developed refined cost models for the operational system (OS). cooled chambers. Composite tanks were also looked at by at least two Independent cost specialists evaluated the competing companies’ contractors to reduce weight. Hybrid combustion was successfully cost models. The contractors were required to convince the demonstrated using a patented staged-combustion concept to government that they have a high probability of success to continue achieve both combustion stability and performance comparable to through CDR and then on to a successful demonstration flight [2]. those of liquid-fuel rockets. Common to all contractors was the Figure 10 shows a concept summary description of each of the four objective of low-cost operations, which drove each one of them to phase IIa concepts. find innovative ways of streamlining their manufacturing, The goals for phase IIb are to develop and evolve the vehicle integration, transporting, storing, and CONOPS processes. designs through CDR. During phase IIc, final flight hardware would Another observation is that, to different degrees, all of the vehicles then be built up and flown no later than fiscal year (FY) 2008. studied were modular and easily scalable. These attributes Demonstration flights are expected to carry prototype autonomous potentially make some of these subsystems great candidates for flight safety systems (AFSS) and low-cost tracking and data relay replacement strap-ons for the Atlas or Delta vehicle families. Also, satellite system (TDRSS) transceivers (LCT2). After the FALCON scaled versions of some of the engines could potentially be used as program is completed, DARPA hands over the demonstration upper-stage engines. FALCON promises to be the beginning of a vehicle system aspects to the Air Force Space Command for new approach to obtain low-cost versatile space launch vehicles from operational system development and implementation. It is possible inception of the design. At the time this paper was completed, only that the vehicle contractor(s) can contract directly with NASA or two FALCON phase IIa contractors, Air Launch and Space X, have private entities (e.g., academia, amateurs, and other government been selected by DARPA to go forward to the phase IIb flight agencies) for commercial launches [2]. demonstration. FALCON is the first concrete program towards the realization of Downloaded by Univ of Wisconsin-Madison on October 8, 2012 | http://arc.aiaa.org DOI: 10.2514/1.23257 affordable and responsive space lift. Each of the four phase IIa VII. Potential New Space Propulsion Systems A. Boost Engines Four competing booster engine design and development programs Air Launch Lockheed were initiated and funded under the Space Launch Initiative (SLI) “Quick Reach” Martin Michoud Program at the Marshall Space Flight Center in 2001. Boeing Liquid Hybrid Pump-Fed Rocketdyne (now Pratt and Whitney Rocketdyne) had two of the Pressure-Fed LO /HTPB Innovative LO /Propane 2 designs, designated as the RS-83 (660,000-lbf (299,371-kgf) thrust 2 CONOPS Modular  6 C-17 LO2=LH2) and the RS-84 [1:05 10 -lbf (476,272-kgf) thrust Launched LO2=RP-1]. TRW Space and Electronics (now Northrop and  6 Grumman) offered a LO2=kerosene-fueled engine in the 1 10 -lb Space X (453,592-kgf) thrust class, which they named the TR-107. The Microcosm “Falcon I” cooptimized for reusable applications (COBRA) was a “Eagle” Liquid Pump-Fed LO2=hydrogen-fueled engine in the 600,000-lb (272,155-kgf) thrust Liquid Pressure-Fed LO /Kerosene High class that was to be designed and developed by a joint venture LO /Jet-A Common 2 2 Reliability/Low between Pratt and Whitney and Aerojet. Pods-Low Cost Cost versus Maximum 1. RS-83 LO =LH Engine Performance 2 2 The RS-83 was a staged combustion LO2=LH2 booster engine Fig. 10 Vehicle characteristics for the four contractors in phase IIa for system with a single fuel-rich preburner in place of the dual the DARPA FALCON program [2]. individual preburners used on the SSME. Taking lessons learned 1324 SACKHEIM

Table 15 RS-83 engine characteristics Table 16 COBRA engine characteristics

Thrust, sea level 663,800 lbf (301,000 kgf) Characteristic Value Thrust, vacuum 749,600 lbf (340,000 kgf) Propellant type H O I 445.7 s vacuum 2 2 sp Mixture ratio 6 Chamber pressure 2800 psia (19,305 kPa) Vacuum thrust 600,000 lb (272,000 kgf) Mixture ratio 6.9 S.L/6 Alt Sea-level thrust 492,590 lb (223,435 kgf) Engine T/W 55 S.L. Vacuum I 454.7 s Area ratio 40:1 sp Sea level I 373.3 s MTBOH/life 50/100 missions sp Chamber pressure 3000 psia (20,684 kPa) Catastrophic reliability 0.999958 Thrust/weight (vac) 75 Throttling 50–100% thrust Engine length 180 in. (457 cm)

from the first-generation reusable launch engine, SSME, which is also manufactured by Rocketdyne, the RS-83 engine was designed to be simpler to build and maintain and to have improved controllability Table 17 RS-84 engine characteristics and increased reliability. Advanced design features included Propellants LO2=RP-1 turbopumps with easy access and fabrication techniques using Thrust, sea level 1:64  106 lbf (483,000 kgf) selectively net-shaped components made through the technology of Thrust, vacuum 1:13  106 lbf (513,000 kgf) powder metallurgy. The engine design, performance, and operating Isp 324 s vacuum characteristics are summarized in Table 15. Chamber pressure 2800 psia (19,305 kPa) The program plan focused on the early development of critical Mixture ratio 2.7 engine components with the overall goal of identifying and reducing Area ratio 20 the risk associated with the development and testing of these Life 100 missions – elements. The engine design team identified five critical component Throttling 65 100% thrust risk-reduction tasks: 1) hydrogen compatible materials, 2) turbine damping, 3) subscale liquid preburner, 4) electromechanical actuator (EMA) sector ball valve, and 5) integrated vehicle health monitoring safety, reliability, and affordability of next-generation reusable space (IVHM) safety/prognostic algorithms. launch and transportation vehicles. The contract specified that a high- Unfortunately the RS-83 design and development program was pressure ORSC cycle was to be used with LO and RP. The engine canceled by NASA along with the other three large reusable booster 2 thrust would have been 1  106 lbf (453,592 kgf). engine programs, when NASA redirected new program goals toward the vision for space exploration in 2004. In its earliest concept phase, the TR-107 had a central pintle injector for both the MCC and the LO2-rich preburner. However, performance and risk analyses soon indicated that the main injector 2. COBRA LO2=LH2 Engine should evolve to a distributed coaxial multielement design, given The goal of the COBRA engine program was to produce a rocket that a single ORSC preburner would be used to drive both the fuel engine prototype that would be simple to operate, provide high and the oxidizer turbopumps. The oxygen-rich preburner (basically a reliability with long life and reduce cost through multiple flight gas generator) retained the pintle injector because its size was within operations. COBRA planned to incorporate a reusable, hydrogen- the pintle injector LO2-RP test database, and because Northrop fueled liquid booster engine with a thrust level of 600,000 lbf Grumman Corporation (NGC) wanted to retain the inherent stability (272,155 kgf). The COBRA engine design consisted of a single fuel- characteristics of the pintle injector. They also wanted the flexibility rich preburner, staged-combustion engine using LO2 and LH2 as to make the preburner throttling for mission flexibility and future propellants. The engine was to be designed to provide a 100-mission growth capability. life span with a 50-mission maintenance checkup interval. The The TR-107 engine was one of several SLI candidate engines proposed engine key design features are given in Table 16. (some of which were described previously) that could be used to Unfortunately this design concept also became a victim of NASA provide primary propulsion for the ETO stage of future reusable program redirection (i.e., 2nd SLI/NGLT cancellation). launch vehicles. Five of the primary technology objectives of this program were accomplished before the SLI efforts were terminated: 1) successful demonstration of a duct-cooled chamber, which 3. RS-84 LO2=RP-1 Engine Downloaded by Univ of Wisconsin-Madison on October 8, 2012 | http://arc.aiaa.org DOI: 10.2514/1.23257 eliminates the need for conventional cooling channels; 2) successful The RS-84 engine program was proposed as the first U.S. reusable demonstration of the preburner pintle injector and propellant mixing, hydrocarbon-fueled, oxygen-rich staged-combustion liquid rocket which improves stable throttling and enables slow controlled startup engine. One of the primary goals of the engine development effort transient performance; 3) establishment of material properties for was to develop a highly reliable and low-cost maintenance engine for oxygen-rich compatible materials, which eliminates the need for the next-generation reusable launch system. The use of kerosene additional coatings and liners; 4) incorporation of mature results in a smaller fuel-tank volume to permit greater propulsive combustion devices that minimize parts count for greater reliability force than other technologies. That benefit translates to more and operability; and 5) systems engineered design optimization to compact engine systems, easier fuel handling and loading on the minimize cycle pressures, which provides a margin to increase ground, and shorter turnaround time between launches. All these engine life. gains, in turn, reduce the overall cost of launch operations, making The central pintle injector technology is also used in the engines routine space flight cheaper and more attractive to commercial for the FALCON program by Space X (LO =RP) and by Air Launch enterprises. In addition, because it is not a cryogenic (or extremely 2 (LO =propane). It has been used in every NGC (TRW) in-space cold) fuel like hydrogen, the propulsion system does not require 2 bipropellant maneuvering and attitude control (AC) thruster for insulation for propulsion-related ducts, valves, lines, and actuators— dozens of major flight satellite programs. Because it can operate with saving weight and cost. Table 17 shows the proposed attributes of the nearly any propellant combination, engines incorporating it are RS-84 engine development effort. Again, this program was canceled excellent candidates for either the reusable booster or second stage of along with the others. affordable responsive spacelift (ARES) or future ORS using gas- generator-driven propellant line pumps. In the case of a reusable 4. TR-107 LO2=RP-1 Engine lifting-body fly-back first-stage vehicle, there could be advantages A primary goal for NASA’s SLI contract for the TR-107 program for a set of throttlable motors along the rear closure of the lifting was to continue development of an engine that could increase the body. Redundant low-risk throttling gas-generator-driven line SACKHEIM 1325

turbopumps would be located within the aft region of the lifting Table 18 XRS-2200 engine characteristics body. Sea-level thrust 206,800 lbf (93,803 kgf) Isp 332 s at 100% & 5.5 MR – 5. XRS-2200 LO =LH Engine Mixture ratio 4.5 6.0 2 2 Chamber pressure 830 psia (5723 kPa) The X-33 program, which began in July 1996, was a 1=2-scale Throttling 57–100 % ’ prototype of Lockheed Martin s proposed single-stage-to-orbit Differential throttling 15% (SSTO) concept named the VentureStar. The program was set up as a Dimensions forward end 134-in. wide  90-in. long (340 cm  229 cm) unique cooperative agreement with Rocketdyne as the supplier of the Aft end 42-in. wide  90-in. long (107 cm  229 cm) XRS-2200 linear aerospike engine. Two of these engines were to be Forward to aft 90 in. (229 cm) used to power the X-33 on suborbital flights to demonstrate the technology needed to proceed with the full-scale VentureStar. ’ The X-33 s aerospike engine, as shown in Fig. 11, was a LO2=LH2 “full-up” rocket engine (prototype level) to be demonstrated in a test- gas-generator cycle engine. Each engine had a single oxidizer bed facility. Senior MSFC management believed that there were a turbopump, fuel turbopump, gas generator, combustion wave number of bright and capable young rocket propulsion engineers that ignition (CWI) system, two aerospike nozzle ramps, 10 thrust had joined the MSFC team but, in a decade, had never participated in, chambers (thrusters) per ramp, two redundant engine controller or even seen a real rocket engine hardware development effort. It was digital interface units (ECDIU), and associated plumbing, valves, felt that a test-bed/prototype engine design and hardware project, and EMAs. The X-33 required two engines in order to provide the conducted entirely inhouse, would be very effective in training and needed thrust vector control (TVC) authority. Table 18 summarizes preparing these bright young engineers (under effective guidance of the XRS-2200 key engine characteristics. A picture of the XRS-2200 a few MSFC experienced/senior rocket development engineers) to be engine being fired at the Stennis Space Center test stand A-1 is shown smart customers for future NASA rocket development projects and in Fig. 12. flight hardware procurement activities. The development philosophy of the X-33 program was to accept The project was named the Fastrac Engine and conducted increased risk to achieve lower costs and a quicker schedule. To do according to all the NASA design and development ground rules and so, the XRS-2200 program relied heavily on the experience gained project criteria. However, just before the formal Fastrac PDR, the from Rocketdyne’s testing of a linear test-bed engine from 1970 to engine requirements were changed in a major way so that it could be a 1972. Where possible, they used existing hardware and designs. The candidate for use on the reusable X-34 Pathfinder/X-Vehicle turbopumps and gas generator were based on J-2 and J-2S engines. demonstrator program. The name of the engine was then changed Component testing was used for design development, proving from Fastrac to the MC-1. Eventually, the X-34 flight demonstrator margins, and for qualifications. Software was tested with hardware in program was canceled by NASA, and all further work of any kind on the loop. Single-engine testing on Stennis Space Center’s (SSCs) A- the Fastrac/MC-1 engine was terminated. 1 test stand was used to verify the design. The two flight engines, in However, some very good technology was developed on this their dual-engine configuration, had a short ignition test and was Fastrac engine project, the most notable of which was the low-cost about to begin acceptance testing when NASA decided not to renew Barber Nichols fuel and oxidizer pump assembly. These pumps were its involvement in the cooperative agreement, because of well along in their development and readiness for incorporation in cancellation of the X-33/VentureStar (Lockheed cancellation) the flight MC-1 engine. When the inhouse MSFC project was programs. terminated, the turbopump assembly elements eventually evolved into being used for two small, low-cost launch-vehicle development projects then being jointly sponsored by DARPA, the USAF (SMC), 6. Fastrac and MC-1 Engines and NASA. Those two projects were intended for use on the In the late 1990s, an in-house rocket engine design and FALCON small launch vehicle program, described in Sec. VI. development project was initiated at NASA/MSFC to give the Barber Nichols TPA elements are currently being used by the younger newly hired propulsion engineers some real hands-on Space X version of FALCON to pump propellants to their 75,000 lbf hardware and design experience. The objectives of this project were to design, build, test, and evaluate a 60,000 lbf LO2=RP-1 low-cost Downloaded by Univ of Wisconsin-Madison on October 8, 2012 | http://arc.aiaa.org DOI: 10.2514/1.23257

Fig. 11 X-33 aerospike engine. Fig. 12 XRS-2200 aerospike engine. 1326 SACKHEIM

fi fi (34,019 kgf) LO2=RP -1 Merlin rst-stage liquid booster engine until the root cause was identi ed and corrected, which could take as [operating at 850 psia (5860 kPa)], and would be used by Lockheed much as a year or more. Although the probability of this is not high, Martin Michoud to pump liquid oxygen to the first and second stages nonetheless it is not zero. In a time of crisis, this could be extremely of their hybrid rocket motor for their version of a low-cost small debilitating to the U.S. national defense needs, if neither of these launch vehicle. systems were available for launch. Therefore it is in the nation’s best interest to develop a second system for the upper stage of these 7. IPD: Integrated Powerhead Demonstrator vehicles. Several organizations have recommended that a new second-stage The IPD engine is a liquid oxygen, liquid hydrogen-fueled full – fl engine ought to be developed in the thrust class of 50,000 100,000 lb ow staged-combustion reusable technology demonstrator engine. (22,680–45,359 kg). Industry has responded to the perceived need The engine development program was begun by the Air Force in with Pratt and Whitney developing the RL-60, Rocketdyne, the MB- 1994 and more recently has been jointly funded by the Air Force 60, Aerojet the AJ-60, and NGC also has an upper-stage engine Research Laboratory and the NASA Marshall Space Flight Center, technology (USET) funded program to develop a 40,000-lb (18,144- with Pratt and Whitney Rocketdyne and Aerojet as the prime kg) LH2 engine. All of these are in various stages of development. contractors, and with testing conducted at the NASA Stennis Space The MB-60 has components with a technology readiness level (TRL) Center. between 6 and 9 depending on the component. The new RL-60 The IPD was intended to demonstrate several new technologies, upper-stage engine (LO =LH ) is under some level of development including high-performance, long-life technologies and materials for 2 2 – fi as Pratt and Whitney teamed with several international partners. reuseable applications and a gas gas injection MCC (for the rst – fi fl Volvo is producing the nozzle while Ishikawajima Harima Heavy time). It is the rst full ow staged-combustion (FFSC) cycle engine Industry (IHI) is providing the hydrogen turbopump. to be developed in the United States. In a FFSC engine, all of the fl Aerojet has worked on the design of their AJ-60 concept but has propellants ow through oxygen-rich and fuel-rich preburners, yet to develop the hardware. They are, however, currently through the turbopumps and into the main chamber, making 100% of developing more physics-based models that are necessary to make the energy of the propellants available to produce thrust. Because all fl real progress in a virtual engine design process. Such a process could of the propellants ow through the turbines, it is possible to trade the have huge leverage in controlling the overall risks of implementing a increased mass flow for lower turbine temperatures, thus reducing full AJ-60 development. the thermal stress on the turbine and increasing life and reliability. All of these options offer various degrees of higher thrust, higher Also, since the hydrogen turbopump is powered by the hydrogen- performance, and higher reliability capabilities than the RL-10 rich preburner and the oxygen turbopump is powered by the oxygen- engine. If heavier payloads are to be placed into higher orbits, this rich preburner, the catastrophic consequences of an interpropellant additional capability will be needed, providing yet another reason to seal leak are avoided. The hydrogen-rich and oxygen-rich gases from develop a new upper-stage engine for this application. The engine the preburners are also used to pressurize their respective tanks, options that appear to be most suitable for the USAF and DoD eliminating the need to use a heat exchanger and eliminating another missions are shown in Table 19. catastrophic failure mode. Hydrostatic bearings in the oxygen and The Atlas V and Delta IV programs both rely on a single-engine hydrogen turbopumps are another important technology demon- fl design for the upper stage (the Pratt and Whitney RL-10). This results strated by the IPD. Hydrostatic bearings are supported by the uid in the risk of a single-point failure for both of these launch vehicles, they pump (except during startup and shutdown), which simplifies and the potential to interrupt the ability to launch heavy payloads for manufacturability and virtually eliminates friction and wear. as much as one to two years if such a failure should occur. A second The IPD is a ground-based demonstrator with a thrust of and alternative engine design, fully tested and certified for 250,000 lbf, mixture ratio 6, 2:1 throttle ratio, and a truncated nozzle. operational readiness, is required to correct this deficiency. In Components of the engine are designed for a lifetime of 200 addition, certain missions will likely require a higher-thrust level missions. than that which is currently available, providing even more reason to develop a second engine option. B. Upper-Stage Engines However, one must notice that most of the engine designs The Pratt and Whitney RL-10A and RL-10B family of upper-stage identified above are not in full-funded development. The engines is now more than 40 years old, and although numerous development of a new rocket engine or motor is a very expensive upgrades have been incorporated over the life of the engine, much of proposition, costing between $100 million and $200 million, or even the design is now outdated and significant improvements in more if problems are encountered during development. In this time of performance and reliability could be achieved with a new design. few launches and a surfeit of competitive launch vehicles, the

Downloaded by Univ of Wisconsin-Madison on October 8, 2012 | http://arc.aiaa.org DOI: 10.2514/1.23257 Furthermore, the engine is the sole available candidate for both development of a new engine is a very risky business. As a result, Atlas V and Delta VI, making the USAF totally dependent on this because of the high cost and active competition, the rocket engine for all large payload launches. propulsion industry has been loath to invest the funds in what they Development and qualification of a new engine to replace the RL- view as a high-risk venture. The engine concepts above have many 10 would be expensive, but given the number of failures in recent technology improvements designed in, but not validated. Many will years, it would seem to be worthwhile to develop an alternative for probably not see much further development without significant this dependence on a single-engine option. The RL-10 has evolved government funding. That picture could become very different if the significantly over its 45 years of service. It began in 1961 with a nNation commits to a serious well-funded long-term program for a vacuum thrust of approximately 15,000 lb (6804 kg) for the RL-10A- new family of responsive space lift vehicles to support a major new 1. Throughout a series of modifications the thrust evolved to an total capability in-space architecture. average level of 24,750 lb (11,226 kg) in the RL-10B-2. Unfortunately this has reduced the margins somewhat, but has resulted in a very useful engine that has probably had every possible C. Strap-On Booster Technologies ounce of thrust rung out of it. Any further development will probably One of the most effective ways to upgrade the payload capability have to come from a new engine design. of launch vehicles is to add strap-ons to the first stage. Solid strap-ons Currently, the EELVs have only one upper stage, some version of have been used frequently for this purpose, but liquids could also be the RL-10. The Delta IV of Boeing uses the RL-10B-2, while the employed. Some of the liquid and hybrid-propellant boosters Atlas V of Lockheed Martin uses the RL-10A-4-1 or -2. What this currently being developed by the various FALCON program means is that both of the major launch systems in use today by the contractors are of an appropriate thrust level that could be a low-cost USAF and other DoD systems are dependent on the same second- alternative to solids for this purpose. Both new solid booster stage engine. Should a failure occur that involves the second-stage technologies in work under IHPRPT, and possibly leveraging liquid engine, all launches with these engines would probably be frozen and hybrid-propellant booster concepts that might be developed SACKHEIM 1327

Table 19 Upper-stage engine options

Pratt & Whitney RL-10A-4 Pratt & Whitney RL-10 Rocketdyne MB-60 Northrop–Grumman TR-40 Aerojet AJ-60 Thrust, lbf (kgf)  1000 23.3 (10.5) 60 (27.2) 60 (27.2) 40 (18) 60 (27.2) Isp, s 451 460 467 461 Mixture ratio 5.5 5.8 5.4 Chamber pressure psia, kPa 620 (4275) 1250 (8618) 2000 (13,789) 1800 (12,410) Weight, lb (kg) 375 (170) 1200 (544) 1300 (590) Area ratio 84 285 300 250 Cycle Closed expander Closed expander Expander bleed Split expander Expander Status Production Preliminary testing Component testing Paper Paper

Table 20 Typical performance characteristics of monopropellant hydrazine thrusters

Thrust ranges 0.025–125 lbf (0.011–57 kgf) – Isp range 225 239 lbf-s=lbm Restart capability 750,000 starts >50 lbf (23 kgf) Pressure operating range 350 psia blowdown <100 psia (0.69 kPa) Radiative thermal control Heaters, various types of heat shields and thermal shunts, as needed

for a new USAF space lift vehicle, should be studied for this 2. Bipropellant Engines application. Key features and capabilities of bipropellant engines currently Lockheed Martin Space Systems has worked on hybrid propulsion available in the U.S. are summarized in Tables 21 and 22. Radiative technologies since 1989. Their initial studies were focused on thermal control is generally used in this thrust class. An innovative replacing the solid rocket boosters on the space shuttle after the bipropellant thruster designated as a secondary combustion Challenger disaster. They worked with the American Rocket augmented thruster (SCAT), has recently been flight qualified and Company (AMROC) and during the DM-01, DM-02, and hybrid flown. This thruster operates in the bipropellant mode on technology operational program (HyTOP) motor developments, MON=N2H4 until the oxidizer is expended and then operates as a which eventually led to the NASA/DARPA hybrid propulsion monopropellant thruster until all fuel is expended. The engine has a development program (HPDP). Within the HPDP, Lockheed Martin thrust range of 5–15 lbf (2.3–6.8 kgf) and is regeneratively cooled tested numerous technologies and increased the TRLs. Under the with a bipropellant Isp max of 326 lbf-s=lbm. MON/MMH liquid DARPA/USAF FALCON program, a number of tests were apogee engines are typically used in combination with low-thrust performed to demonstrate stable hybrid rocket performance. The MON/MMH thrusters used for on-orbit propulsive functions. largest hybrid motor tested to date using the staged-combustion MON=N2H4 liquid apogee engines are advantageous for spacecraft system was the HPDP 250,000-lbf (113,398-kgf) motor, which had a dual-mode propulsion systems that use monopropellant hydrazine 72-in: (2 m) diam and was 30 ft (9 m) long. The motor was catalytic thrusters for ACS and electrothermal hydrazine, or tested 3 times for a total burn duration of 80 s. These tests hydrazine arcjet, thrusters for on-orbit velocity control propulsive demonstrated that the system could be successfully scaled to higher- functions. thrust motors that could potentially be used for booster or first-stage The manufacturers of monopropellant hydrazine and low- and applications. high-thrust bipropellant engines in the U.S. are Aerojet, Redmond, WA; American Pacific, Niagara Falls, NY; and NGC Space Division, Redondo Beach, CA. VIII. Current Status of Spacecraft Onboard Propulsion Technology B. Electric Propulsion A. Chemical Propulsion 1. Electric Powered Propulsion Systems Conventional chemical liquid-propellant propulsion systems in The expanding range of spacecraft size, power availability, and

Downloaded by Univ of Wisconsin-Madison on October 8, 2012 | http://arc.aiaa.org DOI: 10.2514/1.23257 use today are either monopropellant or bipropellant. Liquid- changes in the commercial spacecraft industry environment bipropellant systems are higher performers, but are more complex. presented new challenges to the chemical propulsion community. Monopropellant systems provide a single propellant for There was a clear need to derive higher performance propellants and/ decomposition at the catalyst bed of the combustion chamber. or thrusters. The advent of power-rich spacecraft architectures Widely used, highly reliable, state-of-the art chemical systems are monopropellant hydrazine (N2H4) and bipropellant propulsion systems such as mixed oxides of nitrogen (MON) and Table 21 MON/MMH bipropellant thrusters, low thrust monomethylhydrazine (MMH). For orbit circularization and station Thrust range 0.4–5 lbf (0.18–2.27 kgf) acquisition, dual-mode bipropellant systems using MON=N H are – 2 4 Isp range 250 295 lbf-s=lbm also in use. Restart capability Multiple Pressure operating range 350 psia blowdown <100 psia (0.69 kPa) Radiative thermal control Heat shields, heaters, dog houses 1. Monopropellant Engines Hydrazine thrusters are typically used on satellites or other MON=N H spacecraft to provide either altitude (angular) control or changes in Table 22 MON/MMH and 2 4 bipropellant thrusters, high linear velocity (V). A system which provides attitude control is thrust (liquid apogee engines) known as either ACS (attitude control system) or RCS. Some Thrust range 100–110 lbf (45.4–49.9 kgf) vehicles employ both ACS and V thrusters. Generally, ACS – thrusters tend to be low thrust ( 5 lbf), and V thrusters tend to be Isp range 305 326 lbf-s=lbm – Restart capability Multiple in the higher-thrust range (10 100 lbf). Key capabilities of Engine inlet operating pressure 250 psia (1724 kPa) monopropellant thrusters available in the U.S. are summarized in Radiative/film thermal control Heat shields, heaters, dog houses Table 20. 1328 SACKHEIM

provided the opportunity to take advantage of propulsion options that Table 23 Typical parameters of the 25-cm XIPS thruster provide both high power and high I . Reducing the onboard sp Low-power High-power propulsion-system wet mass requirement can either decrease station keeping orbit raising spacecraft mass or increase payload capability. In addition, greater demands can be placed on the onboard propulsion system, including Active grid diameter, in. (cm) 9.8 (25) 25 increased repositioning or longer duration orbit maintenance— Average Isp, s 3400 3500 increasing useful life. Another option of reduced propulsion-system Thrust, lbf (mN) 0.02 (79) 0.04 (165) Total power consumed, kW 2.2 4.5 wet mass might be a step down to a lower weight-class launch Mass-utilization efficiency, % 80 82 vehicle. These performance enhancements targeted by commercial Typical electrical efficiency, % 87 87 satellite owners are also desirable for military satellites. The propulsion industry accepted these challenges and was readily transitioned to electric propulsion where appropriate and useful. The extent to which the commercial satellite industry has embraced summarized in Table 23. High confidence in the use of the XIPS was electric propulsion is well evident by the simple observation that clearly established by the highly successful flight demonstration there are about 200 operational satellites using electric-propulsion program of the N-STAR ion thruster on the “new millennium systems as of early 2006. sponsored” deep space spacecraft. The state of the art on the XIPS electric-propulsion system is described in [3]. The Boeing 702 spacecraft has a chemical 2. Electrothermal Thrusters propulsion liquid apogee engine, but the use of the high-power mode Starting with the implementation of the electrothermal hydrazine of XIPS in the obit insertion phase greatly reduces the wet mass thruster first by TRW (NGC) on Intelsat V and then by RCA carried by the spacecraft for this task. The high Isp of the ion engines AstroElectronics (now Lockheed Martin) communication satellites, – fi  – for north south station keeping is an additional large savings in a signi cant Isp improvement ( 28%) from 225 to 295=300 s (0.43 propulsion-system wet mass over on-orbit systems, which use 0.50 kW power) was achieved with these thrusters, which electrically monopropellants or bipropellants for this function. heat the decomposition products of monopropellant hydrazine to The military Gapfiller satellite, which has a flight date of late 2006, higher chamber temperatures. Without the complexity of carrying an will use the BSS 702 9.8-in. (25-cm) version of XIPS. Aerojet also oxidizer onboard, this Isp is competitive with low-thrust bipropellant has a major development effort in the ion thruster system technology. – – systems of 2.2 5 lbf (10 22 N), which provide an Isp of 295=300 s Aerojet is completing the thruster, propellant management, and digital control system designs on the 0.5–7.5 kW NASA’s 3. Arcjet Thrusters evolutionary xenon thruster (NEXT) based on an improved version In the early 1990s, Lockheed used the Aerojet MR-509 hydrazine of the N-Star XIPS, led by the NASA Glenn Research Center. Boeing arcjet system (1.8 kW power level, I of 502 s on its series 7000 is developing the power processor. The NEXT system, when sp fi fi satellites). The arcjet continues to evolve with the latest Lockheed quali ed, will provide signi cantly expanded capability for satellite bus, the A2100 satellites, which use the MR-510 arcjet Discovery-class solar electric-propulsion missions. The 15.75-in. – (40-cm) NEXT ion thruster will also be available for other spacecraft system (2.2 kW, 582 s nominal Isp thrusters) for north south station keeping. This thruster takes advantage of the higher satellite power applications. available to substantially increase performance while maintaining the simplicity of a single propellant for onboard propulsive 5. Hall-Effect Thrusters functions. The Russians, in 1971, flew the first Hall-effect thruster, or HET (sometimes identified as a stationary plasma thruster, SPT) on the 4. Ion Thrusters METEOR spacecraft. Over the next two decades, several dozen In the late 1990s, the Hughes Space Division (now Boeing 0.66-kW APT-70 thrusters were used operationally in space [4]. Electrodynamics) successfully introduced the xenon ion propulsion The use of Hall thrusters for satellite north–south station keeping system (XIPS), a xenon propellant girded ion thruster on its BSS 702 promises great savings in wet mass over mono- or bipropellant commercial communications GEO (geosynchronous Earth orbit) chemical propulsion systems. An overview of the underlying physics satellites. The thruster consists of a discharge hollow cathode, three- involved in a Hall thruster is available in [5]. A typical propellant for ring magnetic cusp confinement, three-grid accelerator, and a Hall thruster is a high molecular weight inert gas such as xenon. A neutralizer hollow cathode. The three-grid accelerator used in the power processor is used to generate an electrical discharge between a

Downloaded by Univ of Wisconsin-Madison on October 8, 2012 | http://arc.aiaa.org DOI: 10.2514/1.23257 9.8 in. (25 cm) thruster uses shaped molybdenum grids with cathode and an annular anode through which the majority of 11; 000 apertures to produce the high-perveance (72 pervs at fail propellant is injected. A critical element of the device is the power) xenon ion beam. The XIPS 9.8-in. (25 cm) ion thrusters and incorporation of a radial magnetic field, which serves to impart a spin the associated power supplies xenon power controller (XPCs) to the electrons coming from the cathode and to retard their flow to operate in two modes, 2.2 kW for typical on-orbit functions and the anode. The spinning electrons collide with the neutral xenon, 4.4 kW for augmenting orbit raising. The high-power mode uses ionizing it. The xenon ions are then accelerated electostatically from 4:5kWof bus power to produce a 1.2-kV, 3-A ion beam. The the discharge chamber by the electric potential maintained across the  thruster in this mode produces 165-mN thrust at an Isp of 3500 s. electrodes by the power processor. The velocity of the exiting ions, The high-power mode is used exclusively for the orbit insertion and hence the Isp is governed by the voltage applied to the discharge phase, which greatly reduces the amount of chemical propellant power supply and is typically 49,212–52,493 fps (15; 000– carried by the spacecraft for this task. Nearly continuous operation in 16; 000 m=s) at 300 V. The first flight of a Hall thruster on a U.S. the high-power mode for times of 500–1000 hr is required, spacecraft was in 1998 on STEX, a Naval Research Laboratory depending on the launch vehicle and satellite weight. A low-power spacecraft. In 2004, Loral launched the MBSAT, a geosynchronous mode in which the thruster consumes about 2.2 kW of bus power is satellite, which uses four Faekel SPT-100 Hall thrusters for north– used for the station-keeping function. In the low-power mode, the south station keeping. The performance characteristics of the SPT- beam acceleration voltage is kept the same and the discharge current 100 class of thrusters are shown in Table 24. and gas flow are reduced to generate a 1.2-kV, 1.43-A beam. In this Aerojet Redmond is developing and flight qualifying a 4.5-kW mode, the thruster produces 79 mN of thrust. Because the beam Hall thruster system for the Lockheed Martin build of the USAF voltage remains unchanged for the high-power mode and the thruster advanced extra-high-frequency (EHF) satellites. These Hall thrusters fi mass-utilization ef ciency is nearly the same, the Isp degrades only will operate at two thrust levels, a high thrust for partial orbit transfer slightly compared to the high-power mode to 3400 s. Typical and lower thrust for station-keeping requirements. A launch date is performance parameters of the 9.8 in. (25-cm) thruster are projected as the fourth quarter of 2006. SACKHEIM 1329

Table 24 Characteristics of SPT-100 Hall-effect thrusters that has ready-for-flight application Hall thruster hardware. In addition to the dual-thrust 4.5-kW Hall thruster under contract to Propellant Xenon Lockheed Martin for the EHF spacecraft, Aerojet has a 2.2-kW Hall Thrust, lbf (mN) 0.02 (80) fl Power, kW 1.35 thruster ight prototype unit fabricated and tested. Busek is developing low-watt Hall thrusters under IHPRPT. These are Isp, s 1600 Efficiency, % 50 discussed under that heading. Excellent research and development Life, h >7000 on ion and Hall thrusters and power conditioning units are being conducted at NASA Glenn Research Center, but there needs to be more transfer of technology to companies who can build the product.

The availability of flight-qualified, flight-proven ion and Hall C. Micropropulsion Systems thrusters can be expected to increase the manifesting of these The need for precise positioning of multiple small (micro) technologies because of their high Isps when compared to chemical satellites or formation flying requires a propulsion system capable of propulsion systems. Each type of electric powered thruster has its delivering continuous microthrust levels. One thruster type suited for area of applicability. Ion engines can deliver higher Isps and are well small-satellite, low-power applications that could be used for this suited to missions with high Delta V requirements. In addition to function is Aerojet’s pulse plasma thruster (PPT), designated PRS-1. – satellite north south station keeping and partial orbit transfer This thruster was successfully flown on the EO-1 mission. PPTs rely requirements, Hall thrusters would be suitable for applications such on the Lorenz force generated by an arc passing from anode to as Earth transfer missions. In fact, the European Space Agency cathode and the self-induced magnetic fields to accelerate a small (ESA) SMART 1 mission has already successfully implemented this quantity of chlorofluorocarbon propellant. Teflon has been used as technology. SMART 1 is a small lunar orbiter, which was launched the propellant to date. Pulsed electromagnetic thruster systems in September 2003 as an auxiliary payload on the Ariane 5. consist of the accelerating electrodes, energy storage unit, power fi SMART 1 uses a 1.4-kW Hall thruster and reached the rst moon conditioning unit, igniter supply, and propellant feed system. orbit in December 2004. Because of the mass limitation of the During operation, the energy storage capacitor is first charged to spacecraft and the consequent limitation in the electrical power, the between 1 and 2 kV, and the ignition supply is then activated to thruster used on SMART 1 is a scaled down version of the PPS-1350 generate low-density plasma, which permits the energy storage fi thruster developed and quali ed by SNMECA (France) for capacitor to discharge across the face of the Teflon propellant bar. geosynchronous missions. SMART 1 used its thrusters in a variable The peak current level is typically between 2 and 15 kA, and the arc – power mode (450 1220 W) in this application, which serves as a duration is between 5 and 20 s. The pulse cycle can be repeated at a benchmark for other ETO missions using electric propulsion. rate compatible with the available spacecraft power. The propellant To ensure broader application of ion or Hall thrusters, stronger feed system consists of a negator spring, which pushes the solid emphasis needs to be placed on the development, produceability, and Teflon bar against a stop on the anode electrode. The characteristics reliability enhancement of the components of the entire electric- of the PRS-1 pulsed plasma system, flown on the NASA Goddard propulsion subsystem, which includes not only the thruster but also Earth Observing-1 (EO-1) spacecraft, are shown in Table 25. the propellant feed system and the power processing unit (PPU). Two The PPT-1 has demonstrated control of the spacecraft pitch with companies, Moog and Vacco Industries, are leading efforts to the momentum wheels completely disabled, including during image fl produce propellant-management components and systems for ight acquisition with the advanced land imager instrument. On-orbit tests electric-propulsion systems. They are also developing next- have demonstrated no detectable electromagnetic interference with generation designs that will trim the weight of the propellant- the spacecraft, spacecraft communications, or the payload management system. Moog was the supplier of the xenon- instrument. fl propellant-management assembly that is ight operational on the A micro-PPT was developed to provide attitude control on Loral MBSAT. Currently Vacco provides the propellant-manage- FalconSat when it is launched. The characteristics of this thruster are ment system components for the BSS 702 XIPS system. Vacco is in shown in Table 26. Colloid thruster technology also received the process of qualifying highly integrated xenon latch valve considerable attention. Colloid thrusters have demonstrated modules for the Lockheed Martin EHF spacecraft. Historically the remarkable range, precision, and controllability in the micro-newton PPU has been the dominant cost driver for electric-propulsion regime, with single spray sources providing thrust in the 0:3–3 N systems because of the requirement for heavy power converters and range with nano-newton resolution. The response time across the full thermal management systems. Aerojet Redmond designs and builds thrust range is only 2–3 s, achieved via a microvalve, and fine high-power converters to support the electric-propulsion subsystems adjustments, performed by varying acceleration voltages, may be Downloaded by Univ of Wisconsin-Madison on October 8, 2012 | http://arc.aiaa.org DOI: 10.2514/1.23257 they manufacture. They are also working on development of solar- realized in milliseconds. The characteristics of the microthrusters for electric direct drive (i.e., using a high-voltage solar array to provide the NASA ST7 mission are shown in Table 26. power directly to a Hall thruster at voltage levels needed to drive thruster discharge). Qualification of solar-electric direct drive would greatly reduce the cost and weight of a Hall electric-propulsion IX. Current Status of In-Space Propulsion Technology system while reducing array size. Reduction in array size is an added Current and future needs for satellites and in-space vehicles exist savings in spacecraft weight. The potential payoff for direct drive in the following areas: makes this is an extremely worthy goal to pursue. Additional weight 1) Strategic assets for communication, early warning, Earth savings could be obtained with direct drive using power from observation, navigation, reconnaissance, surveillance, and weather; advanced solar arrays now available commercially, such as 2) technology development work in space; ENTECH/ABLE’s solar concentrator arrays with refractive linear 3) responsive space operations. element technology. The latter type array provided the 2.7-kW power All of the U.S. satellites and technology platforms require source for the successful deep space 1 basic mission and its extended propulsion subsystems operating in space to provide the impulse mission to the comet Borrelly in 2001 [6]. ENTECH/ABLE also has necessary to adjust velocity, change orbit altitude, provide attitude available a next-generation array that is a stretched lens array. The control, station keeping, and end-of-life deorbit. These propulsion synergy of coupling flight-proven advanced array technologies with needs are being satisfied currently by using state-of-the-art chemical direct drive for Hall thrusters needs to be explored. propulsion and increasingly, by electric-propulsion subsystems. It To satisfy increasing demand, a larger industrial base is needed should be noted that the “state of the art” has been undergoing major over what now exists for the manufacture of electric-propulsion changes over the past 15 years, and therefore represents a very systems and components. Boeing electrodynamics division and advanced capability in many areas. For kilogram-weight-class Aerojet Redmond appear to be the only commercial sources of girded technology development satellites with unique mission capabilities, ion thrusters and power converters. Aerojet is the only U.S. source micropropulsion systems may be required for all maneuvers other 1330 SACKHEIM

Table 25 PPT performance characteristics [7,8]

Characteristic EO-1 Dawgstar Maximum input power 70 W (one thruster-EO-1 operations)-100 W design 15.6 W (two thrusters; slow charge) 36 W (two thrusters; fast charge) Thrusters/system 2 8 Total system impulse 405 lbf-s (1850 N-s) (EO-1 propel. load) >3372 lbf-s >337 lbf-s (1500 N-s) (15,000 N-s) (system life) Impulse bit 20–193 mlbf-s (90–860 mN-s), throttleable 15 mlbf-s (66 mN-s) Pulse energy 6.3–41 ft-lb (8.5–56 Joules), throttleable 3.6 ft-lb (4.9 Joules) Maximum thrust 193 lbf (860 N) (EO-1); 0.3 mlbf (1.2 mN) (design) 45 mlbf (200 mN) (high-speed mode) – ‡ Isp 650 1,350 s 332 40 s Thrust to power ratio 2:74 mlbf=W (12:3mN=W) 2 mlbf=W (9:7mN=W) Total mass 10.8 lb (4.9 kg) (2 PPTs, a power processing unit, 9.2 lb (4.2 kg) (8 PPTs, a power processing unit, and propellant) and propellant) Propellant PTFE PTFE Propellant mass 0.15 lb (0.07 kg)/thruster (as fueled) 0.15 lb (0.07 kg)/thruster; 1.2 lb (0.56 kg)/system

entail a small fleet of permanently based space maneuvering tugs that Table 26 Summary of small electric thruster capabilities would have no other function [with its own dedicated guidance, Colloid thruster Micropulsed plasma thruster navigation, and control (GNC) and telemetry/command system] than to rapidly maneuver space assets to new stations as required. This Dry mass 6.6 lb (3 kg) 1.54 lb (0.7 kg) fleet of tugs should also be on-orbit refuelable to extend their I 1000 s 800 s sp lifetimes. Min. Ibit 0.022 lbf-s (0.1 N-s) 16.9 lbf (75 N) Power 25 W 10 W Of course, some combination of the above approaches for slow or Delta V 181 fps (360 m=s) 1230 fps (375 m=s) rapid maneuvering and repositioning could also be adopted for TRL 7 7 operational use to provide more robust, flexible, survivable, and Demo(s) NASA ST7 FalconSat–3 long-life capabilities.

A. Chemical Propulsion Work Under IHPRPT than rapid inclination change. A summary of the in-space rocket As discussed earlier, the USAF has established a program referred engines and thrusters in use today has already been presented in to as IHPRPT. This program is a joint government and industry effort Table 1. focused on developing technologies for military global reach The nation will continue to need higher performance in-space capability, strategic missiles, long-life or spacecraft capability, and propulsion technologies for military satellites, commercial space tactical missile capability. The performance metric increases desired systems, and civil space vehicles of various types for various for spacecraft propulsion are shown in Table 27. missions. There are a number of approaches for meeting these needs. For large Earth orbiting assets, for example, one could use an 1. Alternative Propellant Developments for Liquid-Propellant Engines fi onboard, low-thrust very high fuel ef ciency performance system Research and development is under way in house at AFRL and in fi such as a Hall-effect thruster that would re continuously to complete industry on several energetic monopropellants for potential use for – a large station change or repositioning maneuver at high Isp (2000 orbit circularization, orbit altitude/position changes, fly-out and 3000 s). However, such maneuvers for large assets are measured in maneuvering, and attitude control to meet the phase II IPRPHT goals velocity changes of feet per second per hour and take weeks to travel (Table 28). Hydroxy ammonium nitrate (HAN) and AF-315E are 10,000 mile (16,093 km). two of the propellants under study. The main advantages are higher- – A second way would be to use a moderate-thrust (100 200 lbf density impulse than state-of-the-art chemical monopropellant or – (45 91 kgf)), modest-performance chemical propulsion thruster at bipropellant systems and lower propellant toxicity. The industry is – 350 360 s, such as LO2=N2H4 using a cryocooler to keep the LO2 experimenting with various blends to achieve a balance between Downloaded by Univ of Wisconsin-Madison on October 8, 2012 | http://arc.aiaa.org DOI: 10.2514/1.23257 from boiling away. Velocity changes of hundreds of feet per second safety of handling and performance. Phase II potential has not yet depending on the propellant mass available can be achieved in been realized at the thruster level. Theoretical performance minutes to hours for platforms of thousands of pounds, permitting – calculations indicate an Isp of 250 270 s. Measured Isp to date is position changes of thousands of miles per day. With higher-thrust 10% below that level. There are at least three leading technical levels maneuvering times could be quite rapid. challenges associated with HAN-type propellants: 1) selection of One important technology that would permit multiple and longer suitable chamber materials for the thrusters because the high- life maneuvers of critical assets would be to use an on-orbit refueling performance blends tend to run hotter than monopropellant system to resupply the propellants while changing stations. The on- hydrazine, 2) the high temperature needed on the catalyst bed for orbit refueling capability would enable the space asset to stay alive as initial startup, and 3) ignition delay problems. These challenges are long as everything kept working and to make as many rapid station solvable, but the feasibility of large-scale propellant production and changes as required. The capability for autonomous, robotic on-orbit long-term material compatibility are still outstanding. These docking and refueling will be demonstrated with hydrazine and high- “designer” monopropellants may fill a small-niche mission need, pressure helium in space by the end of 2006, by the DARDA, USAF, such as low delta V, small spacecraft applications; however, existing NASA sponsored orbital express (OE). state-of-the-art bipropellant systems are competitive in performance A third way to implement rapid station changes would be to have a with HAN, and MON/MMH systems bring with them an already large (even on-orbit refuelable) space tug with high-performance substantial history of proven performance and reliability. electric propulsion for slow strategic moves or a high-thrust, modest- performance chemical system for responsive maneuvers. Or have some combination of propulsion systems on the tug that would fly up, 2. Combination Thrusters Dual-Mode Capability dock with a key space asset, and move it to the desired new SCAT is the first thruster designed to operate with “bimodal” operational location. The tug could then demate from the spacecraft capability, in either a bipropellant or a monopropellant mode. In its and fly on to reposition other assets as needed. This approach would monopropellant mode, SCAT decomposes hydrazine in a catalyst SACKHEIM 1331

Table 27 Performance metric increases for spacecraft propulsion

Spacecraft propulsion Phase I Phase II Phase III

Improve Itot=mass wet† 20%=200% 35%=500% 75%=1250% (electrostatic/electromagnetic) Improve Isp 5%=10% 10%=15% 20%=20% (bipropellant/solar thermal) Improve density-Isp 30% 50% 70% (monopropellant) Improve mass fraction 15% 25% 35% (solar thermal)

Table 28 Chemical propulsion IPRPHT phase II goals.

Activities Objective Phase II demonstration To demonstrate technologies that enable realization of USAF IHRPT phase II goals with improved Isp, life, thrust to weight and lower cost Liquid engine alternative propellant 1) To develop and demonstrate a catalytic engine for AF-M315E monopropellant that delivers development program (LEAP-DP) IHPRPT phase II performance 2) To design and fabricate a 25-lbf heavyweight workhorse thruster and flight weight thrusters 3) To hot-fire demonstrate flight weight thruster with AF-M315E monopropellant Energetic propellant To develop and demonstrate the scale up of energetic propellant formulations

‡ ‡ bed chamber [9]. The decomposition products (NH3 N2 H2) 3) accomplishing the transfer with minimal venting of propellants flow out through a second small chamber and exit through a and pressurant gases; conventional nozzle with an expansion ratio of about 100:1. The 4) meeting a requirement for zero dribble from quick disconnect fi conventional monopropellant N2H4 thruster has a steady state Isp of ttings; 230 s and can provide thrusts from 0.8 to 4.5 lbf (0.36–2 kgf). This 5) reuse of pressurant gases; system has been under development for years at Northrop-Grumman 6) completely autonomous operations; Corporation and is now flight qualified on several NGC satellites. 7) replacement of deficient electrical boxes/subsystems. In its bipropellant mode, N2H4 is turned on to cool the second A primary requirement is to have no spacecraft contamination chamber, vaporizes, and then combusts the NO2 vapor with the N2H4 occur during or following the transfer. The present plan is to decomposition products in the second chamber to produce an Isp of demonstrate transfer of hydrazine only. Transition to a fully 325 s. Because the second chamber is regen cooled, it can be made operational fluid transfer docking satellite and autonomous servicing of nonrefractory metals such as nickel and L-605 alloy. This provides system could have important benefits for mobility of future USAF essentially unlimited operating life. In its bipropellant mode, the and NASA satellite constellations and for reusable fly-out tugs for thruster can produce 4–14 lbf (1.8–6.35 kgf) thrust. NGC is various in-space missions. Some assets could be launched into orbit developing higher-thrust versions and also exploring other without having to have an onboard propellant capacity to support all propellant combinations. of the missions the space vehicle may be functionally capable of SCAT’S combination of operating modes permits the most carrying out. efficient use of onboard propellant, thereby providing greater mission flexibility. SCAT allows a single engine to burn either hydrazine (monopropellant mode) or hydrazine and N2O4 X. Potential Future Propulsion Needs for Space (bipropellant mode), which means that the oxidizer tank can be Exploration fully depleted and then the fuel tank can be fully depleted. This total ’ “ utilization of available propellant is not possible with any other Enabling the president s vision to extend human presence across bipropellant thruster. SCAT also has a very wide range of allowable the solar system, starting with a human return to the moon by the year Downloaded by Univ of Wisconsin-Madison on October 8, 2012 | http://arc.aiaa.org DOI: 10.2514/1.23257 – 2020, in preparation for human exploration of Mars and other mixture ratios (0.95 1.6), so it becomes much easier to balance ” fl oxidizer-versus-fuel usage over a mission, unlike common LAEs, destinations will require a robust and reliable eet of launch which have very tight limits on operating mixture ratio (e.g., vehicles [10]. With the retirement of the shuttle scheduled for the 1  0:05). It can also operate over a deep or wide blowdown range, year 2010, the United States is now faced with the challenge of 4.5–1, just like catalytic monopropellant hydrazine thrusters. replacing its only human-rated and heavy-lift launch vehicle. To meet this challenge, NASA was elected to pursue an evolutionary shuttle-derived approach to ensure the rapid development of a launch B. On-Orbit Refueling: Orbital Express vehicle for the robotic lunar missions targeted for 2008 and the DARPA is sponsoring the orbital express spacecraft program to human lunar missions targeted for 2015–2020 (i.e., the crew launch demonstrate the practicality of carrying out autonomous/robotic on- vehicle, or CLV). Adding to the difficulty of maintaining U.S. orbit refueling of spacecraft propulsion systems. Reasons for the technical superiority is the new emergence of China as the third DARPA interest in in-space refueling is that some satellites that nation to successfully put a human into space. The United States will retain full functionality have to be retired because the original now have to compete with multiple nations in its quest to remain the propellant load carried into orbit with the satellite is exhausted. If world leader in space exploration. spacecraft can be designed to be refueled in space, they could When NASA initiated the Apollo program in the 1960s, the continue to operate for much greater periods. To date, the United Saturn V was developed with the capability to lift 140 t to LEO. This States has not been successful in performing acceptable robotic vehicle provided the heavy-lift capability that enabled NASA to transfer of fuels in space. The capabilitites to be demonstrated by the complete lunar missions with a single launch. The most recent orbital express include the following: human exploration concept studies indicate that future missions will 1) autonomous rendezvous and docking of two independent require payloads in excess of 100 t to LEO. Currently, the largest spacecrafts; payload capability provided by an existing launch vehicle is the 2) an efficient zero-g propellant transfer pump; space shuttle at 27.5 t. Although shuttle-derived vehicle concepts 1332 SACKHEIM

appear to provide an expedient path to the development of a heavy- development of propellant depots would provide extended service lift launch vehicle, initial vehicle concept studies indicate that life to all reusable in-space propulsion systems and provide NASA expendable or partially expendable systems will provide the most with the option to launch transportation elements without fuel, cost-effective access to space for human exploration missions. The providing additional payload mass margin. Along these lines, largest Delta- and Atlas-class heavy expendable launchers can only introduction of oxygen (and possibly fuel) production from in situ deliver payloads in the 20- to 25-t range. Based on economic factors resources can have a significant impact on reducing mission cost and and the limited number of planned exploration missions, the United mass. For every kilogram delivered to the surface of the moon and States may have to delay the development of a large heavy-lift launch Mars equates to approximately 8.8 lb (4 kg), or more, in LEO. system, which could force NASA to rely on the use of multiple ETO Therefore, in situ propellant production (ISPP) of oxygen, which can launches with on-orbit assembly of the spacecraft and transfer of make up 75% of propellant mass, can significantly reduce mission expendables performed in LEO. cost and mass. Full propellant production has an even greater impact, The selection and sizing of the launch-vehicle fleet will represent a and ISPP also enables reusability of lander assets and can support in- landmark architectural decision in the future exploration of space. space propellant depots and reusable in-space stages for reduced cost The CONOPS developed for the lunar and follow-on Mars trans-Earth transportation. campaigns will dictate the design of each element required for these Using its design approach, the Apollo program developed 10 missions. The sizing of the TLI stage, the trans-Earth injection different engine types specifically designed for each transportation (TEI) stage, and 29 other vehicle subsystems in the notional element on the Saturn V. Although these same transportation architecture shown in Fig. 13 will be some of the primary drivers elements and more will be required for future planetary exploration used to determine the optimum payload size for future launch missions, it is not economically feasible to design, test, and build new vehicles. unique engines for each element. As discussed previously, the only Based on the most recent vehicle trades, it is not feasible, under the U.S.-built liquid-propellant upper-stage-enabling rocket engine is current NASA budget constraints, to pursue a single-launch the RL-10XX oxygen/hydrogen engine, which was initially approach for the deployment of a planetary exploration vehicle. A developed in the late 1950s. Even though the RL-10XX has gone multilaunch approach will be required to enable human exploration through several design modifications to upgrade engine perform- of the moon, Mars, and other planets. The need to assemble an ance, an advanced upper-stage engine with multiple application exploration spacecraft in LEO will necessitate accelerating the capabilities is still identified as a key technology needed for its many development of advanced autonomous rendezvous and docking applications in the human exploration of the moon and Mars. (ARD) systems. These systems will provide NASA with the NASA will conduct the studies and develop the architectures from capability to launch unmanned elements (including propellant) of the which will come the requirements for launch systems, orbit transfer exploration vehicle separately and then execute on-orbit assembly and trajectory insertion vehicles, in-space fluid management and propellant-loading operations in LEO. Advanced ARD systems systems, and the many other propulsion elements needed for human will allow NASA to functional test the fully assembled exploration exploration of the moon and Mars. Some of these requirements will vehicle before launching the crew, which will enhance crew safety be met from work now being done by NASA and the USAF, but and provide time to work around any anomalous conditions that may many of them will require new systems. As we prepare for human occur during the vehicle checkout phase. Advanced ARD systems space exploration, we must recall that the United States no longer has could also be used for station keeping in the event of a component or the base of industrial capability on which Apollo was built. Instead, subsystem malfunction where replacement units would have to be the exploration program will work with an industry shrunk through delivered on a subsequent launch. unrelenting consolidation and losing its more experienced The development of highly efficient in-space propellant- individuals as they age. If the exploration program is to succeed, management systems will provide NASA with the capability for its studies and plans will consider how best to invest both its long-term on-orbit cryogenic storage of propellants and autonomous advanced technology funds and its development funds so as to transfer of stored propellant for in-space refueling operations. The maximize the yield from, and improve, the capability that remains. Downloaded by Univ of Wisconsin-Madison on October 8, 2012 | http://arc.aiaa.org DOI: 10.2514/1.23257

Fig. 13 Notional lunar system and mission scenario. SACKHEIM 1333

XI. Conclusions critically stressed element associated with space flight, particularly fl ’ Robotic and human space missions into and beyond LEO are very human space ight. Yet, just when the nation s need is growing, the complex and risky endeavors. When humans are the primary cargo, national capability on which it can draw is shrinking. Only the space- especially to more distant locations such as the surface of the moon, exploration program can address this problem and is given a charge all space flight operations become much more complex; more whose execution requires that the problem be addressed. To ensure mission success for future exploration missions, the important there is now the element of great danger that must be fi ’ addressed and reconciled. This statement is based on fact, not nation must take strong de nitive steps to retain and enhance today s opinion and politics, and has been demonstrated in many ways over diminishing propulsion technology knowledge and skill base. A the history of space flight worldwide since the middle of the last concerted effort by the government to increase the level of much- century. In the case of human space flight, the subject of safety has to needed new propulsion research and development activities would become the dominant factor and criterion for mission success of any lead to improved performance and an overall reliability increase of kind. The cost of failure in space is enormous. The physics involved future propulsion-system designs, as well as build on the essential lessons that were learned from propulsion systems and associated and the operational environment are extremely marginal and fl unforgiving. successes that were own on actual missions during the earlier Space flight is nothing like airplane/airline flight. In the first place, ICBM, Apollo, and STS eras. airplanes only flyupto20–50,000 ft (6–15,240 m) and then, using the existing atmosphere or some gentle type of thrust reversal (or a Acknowledgments fl combination), gradually reduce the energy required to y from point Special thanks go to G. W. Elverum, Jr. (TRW/ATD, V.P., retired, ’ to point inside the Earth s atmosphere using aerodynamic braking to now private consultant); Y. Brill (private consultant), G. Dressler allow a safe and highly controlled gentle landing. Space vehicles, (Northrop Grumman Space Technology, Chief Engineer for however, must add a minimum of 30,000 fps (9144 m=s) of energy Propulsion) for technical support, and Aerojet, Pratt and Whitney just to reach a minimum safe parking orbit, above most of the drag of Rocketdyne, and Northrop Grumman Space Technology for the use ’ the Earth s atmosphere. The energy required to go much further in of their openly published marketing data and pictures available from – the solar system (e.g., to the moon and beyond) can be 2 3 times that their respective Web sites. Also thanks to John Wiley and Andy of ETO, plus the original 30,000 fps (9144 m=s). Gamble (NASA Marshall Space Flight Center) and the people at the Then, to ensure a gentle landing on the destination planet and/or 10 NASA field centers and headquarters who participated in the back on Earth, all of this energy that was added by extremely space propulsion strategic partnership planning efforts and provided powerful rockets must be reduced back down to nearly zero, relying significant background information for this document. upon frictional heating at hypersonic reentry speeds, greater than Mach 25 at temperatures greater than 5432F (3000C), and then drag to kill off the remaining subsonic velocity with aerodynamics. References This is because all space vehicles (from the surface of the Earth to [1] Mueller, G. E., “‘Apollo’ the First Space Systems of Systems,” Human anywhere in space) are always extremely weight limited, even with Exploration Space Technology Conference, Dec. 2004. the best rocket technology that could be developed in the near future. [2] Sackheim, R. L., London, J. R., III, and Weeks, D. J., “The Future for ” The cost, in terms of GLOW, to carry the full allotment of thermally Small Launch Vehicles, Space Technology and Applications International Forum (STAIF–2005), Albuquerque, NM, Feb. 2005. (high thrust-to-weight) energized propulsion/propellant for all “safe ” [3] Goebel, D. M., Martinez-Lavin, M., Bond, T. A., and King, A. M., braking functions would be enormous, and the amount of payload “Performance of XIPS Electric Propulsion in On-Orbit Station Keeping would be prohibitively small (if not negative). of the Boeing 702 Spacecraft,” AIAA Paper 2002-4348, July 2002. During powered space flight using rocket propulsion, because of [4] Bober, A. S., and Maslennikov, N. A., “SPT in Russia—New the enormous amount of energy that must be released under Achievements,” Proceedings of the 24th International Electric extremely controlled, but nearly marginal and very high-stress Propulsion Conference, Moscow, Sept. 1995, pp. 54–60. conditions, it is this high-power/high-energy release period that is the [5] Kaufman, H. R., “Technology of Closed-Drift Thrusters,” AIAA most dangerous phase (or phases) of the mission. To reinforce this Journal, Vol. 23, No. 1, Jan. 1985, pp. 78–87. [6] Jones, P. A., Murphy, D. M., Allen, D. M., Caveny, L. H., and Piszczor, point it is important to note that the probability of a catastrophic “ 7 fl M. F., SCARLET: A High-Payoff, Near Term Concentrator Solar airplane failure is about 1 in 10 ights, whereas for the space shuttle, ” fl Array, AIAA Paper 96-1021, Feb. 1996. the probability of catastrophic ight during the powered phase is [7] Benson, S. W., Arrington, L. A., and Hoskins, W. A., “Development of 2 about 4 in 10 flights, or about 1 in 100 flights if you consider actual PPT for EO-1 Spacecraft,” AIAA Paper 99-2276, June 1999. results to date. [8] Rayburn, C., Campbell, M., Hoskins, W. A., and Cassady, R. J., It is also interesting to note that from a power plant internal “Development of a Micro Pulsed Plasma Thruster for the Dawgstar Downloaded by Univ of Wisconsin-Madison on October 8, 2012 | http://arc.aiaa.org DOI: 10.2514/1.23257 operation point of view, the horsepower-to-weight ratio of the best jet Nanosatellite,” AIAA Paper 2000-3256, July 2000. engine flying (either the F-22 or the Joint Strike Fighter) is about [9] Sackheim, R. L., Hook, D. L., and Joseph, G. W., “Satellite Propulsion 19:1. This compares with an average horsepower-to-weight ratio for and Power System,” Patent No. 5,572,865, filed 12 Nov. 1996. the 400,000-lbf (181,437 kgf) thrust SSME of about 800:1; and the [10] Bush, G. W., published vision statement, 14 Jan. 2004. initial start transient for the SSME takes the combustion chamber from about 400 to 7000F(240 to 3817C) (in the core) in 1s. D. Talley Therefore, it is the rocket propulsion engine system that is the most Associate Editor This article has been cited by:

1. Akira Kakami, Shinji Beppu, Muneyuki Maiguma, Takeshi Tachibana. 2012. Performance of Dimethyl Ether Arcjet Thrusters and Its Dependence on Electrode Configurations. Journal of Propulsion and Power 28:3, 603-608. [CrossRef] 2. Akira KakamiShinji BeppuMuneyuki MaigumaTakeshi Tachibana. 2012. Performance of Dimethyl Ether Arcjet Thrusters and its Dependence on Electrode Configurations. Journal of Propulsion and Power 28:3, 603-608. [Citation] [PDF] [PDF Plus] 3. Laurent Catoire, Steven D. Chambreau, Ghanshyam L. Vaghjiani. 2012. Chemical kinetics interpretation of hypergolicity of dicyanamide ionic liquid-based systems. Combustion and Flame 159:4, 1759-1768. [CrossRef] 4. Anthony B. Murphy. 2012. Transport coefficients of plasmas in mixtures of nitrogen and hydrogen. Chemical Physics 398, 64-72. [CrossRef] 5. Akira KAKAMI, Shinji BEPPU, Muneyuki MAIGUMA, Takeshi TACHIBANA. 2012. Thrust Evaluation of an Arcjet Thruster Using Dimethyl Ether as a Propellant. TRANSACTIONS OF THE JAPAN SOCIETY FOR AERONAUTICAL AND SPACE SCIENCES 55:2, 116-122. [CrossRef] 6. Richard Strunz, Jeffrey W. Herrmann. 2011. Reliability as an Independent Variable Applied to Liquid Rocket Engine Test Plans. Journal of Propulsion and Power 27:5, 1032-1044. [CrossRef] 7. Richard StrunzJeffrey W. Herrmann. 2011. Reliability as an Independent Variable Applied to Liquid Rocket Engine Hot Fire Test Plans. Journal of Propulsion and Power 27:5, 1032-1044. [Citation] [PDF] [PDF Plus] 8. 2011. Full issue in PDF / Numéro complet en form PDF. Canadian Aeronautics and Space Journal 57:2, 115-153. [CrossRef] 9. Hanafy M. Omar, Moumen Idres, Raed Kafafy. 2011. Particle swarm optimization of a micro air-launch vehicle trajectory. Canadian Aeronautics and Space Journal 57:2, 115-120. [CrossRef] 10. Hai-Bin TangXin-Ai ZhangYu LiuHai-Xing WangChen-Bo ShiBin Cai. 2011. Experimental Study of Startup Characteristics and Performance of Low-Power Arcjet. Journal of Propulsion and Power 27:1, 218-226. [Citation] [PDF] [PDF Plus] 11. Ralph L. McNuttScience Satellites: Interplanetary Spacecraft . [CrossRef] 12. Hai-Xing Wang, Jin-Yue Geng, Xi Chen, Wen Xia Pan, A. B. Murphy. 2010. Modeling Study on the Flow, Heat Transfer and Energy Conversion Characteristics of Low-Power Arc-Heated Hydrogen/Nitrogen Thrusters. Plasma Chemistry and Plasma Processing 30:6, 707-731. [CrossRef] 13. Xi Chen. 2010. The impact force acting on a flat plate exposed normally to a rarefied plasma plume issuing from an annular or circular nozzle. Journal of Physics D: Applied Physics 43:31, 315205. [CrossRef] Downloaded by Univ of Wisconsin-Madison on October 8, 2012 | http://arc.aiaa.org DOI: 10.2514/1.23257